international energy agency
world energy outlook 2oo8 Please note that this PDF is subject to specific restrictions that limit its use and distribution. The terms and conditions are available online at
http://www.iea.org/Textbase/about/copyright.asp
world energy outlook 2008 Are world oil and gas supplies under threat? How could a new international accord on stabilising greenhouse-gas emissions affect global energy markets? World Energy Outlook 2008 answers these and other burning questions. WEO-2008 draws on the experience of another turbulent year in energy markets to provide new energy projections to 2030, region by region and fuel by fuel. It incorporates the latest data and policies. WEO-2008 focuses on two pressing issues facing the energy sector today: n
rospects for oil and gas production: How much oil and gas exists and how much P can be produced? Will investment be adequate? Through field-by-field analysis of production trends at 800 of the world’s largest oilfields, an assessment of the potential for finding and developing new reserves and a bottom-up analysis of upstream costs and investment, WEO-2008 takes a hard look at future global oil and gas supply.
n
ost-2012 climate scenarios: What emissions limits might emerge from current P international negotiations on climate change? What role could cap-and-trade and sectoral approaches play in moving to a low-carbon energy future? Two different scenarios are assessed, one in which the atmospheric concentration of emissions is stabilised at 550 parts per million (ppm) in CO2 equivalent terms and the second at the still more ambitious level of 450 ppm. The implications for energy demand, prices, investment, air pollution and energy security are fully spelt out. This groundbreaking analysis will enable policy makers to distill the key choices as they strive to agree in Copenhagen in 2009 on a post-Kyoto climate framework.
With extensive data, detailed projections and in-depth analysis, WEO-2008 provides invaluable insights into the prospects for the global energy market and what they mean for climate change.
€150 (61 2008 23 1 P1) ISBN: 978-92-64-04560-6
international energy agency
world energy outlook 2oo8
INTERNATIONAL ENERGY AGENCY The International Energy Agency (IEA) is an autonomous body which was established in November 1974 within the framework of the Organisation for Economic Co-operation and Development (OECD) to implement an international energy programme. It carries out a comprehensive programme of energy co-operation among twenty-eight of the OECD thirty member countries. The basic aims of the IEA are:
To maintain and improve systems for coping with oil supply disruptions. To promote rational energy policies in a global context through co-operative relations with non-member countries, industry and international organisations.
To operate a permanent information system on the international oil market. To improve the world’s energy supply and demand structure by developing alternative energy sources and increasing the efficiency of energy use.
To promote international collaboration on energy technology. To assist in the integration of environmental and energy policies. The IEA member countries are: Australia, Austria, Belgium, Canada, Czech Republic, Denmark, Finland, France, Germany, Greece, Hungary, Ireland, Italy, Japan, Republic of Korea, Luxembourg, Netherlands, New Zealand, Norway, Poland, Portugal, Slovak Republic, Spain, Sweden, Switzerland, Turkey, United Kingdom and United States. The European Commission also participates in the work of the IEA.
ORGANISATION FOR ECONOMIC CO-OPERATION AND DEVELOPMENT The OECD is a unique forum where the governments of thirty democracies work together to address the economic, social and environmental challenges of globalisation. The OECD is also at the forefront of efforts to understand and to help governments respond to new developments and concerns, such as corporate governance, the information economy and the challenges of an ageing population. The Organisation provides a setting where governments can compare policy experiences, seek answers to common problems, identify good practice and work to co-ordinate domestic and international policies. The OECD member countries are: Australia, Austria, Belgium, Canada, Czech Republic, Denmark, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Japan, Republic of Korea, Luxembourg, Mexico, Netherlands, New Zealand, Norway, Poland, Portugal, Slovak Republic, Spain, Sweden, Switzerland, Turkey, United Kingdom and United States. The European Commission takes part in the work of the OECD.
© OECD/IEA, 2008 International Energy Agency (IEA), Head of Communication and Information Office, 9 rue de la Fédération, 75739 Paris Cedex 15, France. Please note that this publication is subject to specific restrictions that limit its use and distribution. The terms and conditions are available online at http://www.iea.org/Textbase/about/copyright.asp
FOREWORD
Out of the turmoil of the energy markets of the last 12 months and our evaluation of future influences on the sector has emerged a new underlying price assumption for the World Energy Outlook — an oil price through to 2030 which nudges twice the level in WEO-2007. The era of cheap oil is over. This alone should be enough to make policy makers sit up. But so should the results set out in this volume. On present trends, just to replace the oil reserves that will be exhausted and to meet the growth in demand, between now and 2030 we will need 64 mb/d of new oil-production capacity, six times the size of Saudi Arabia’s capacity today. The big resources lie, increasingly, in a few countries, whose share of the world market climbs inexorably. The international oil companies have diminishing access to these reserves. Investment per barrel of oil produced has shot up. But big producers are becoming so wealthy that they are losing the incentive to invest. Though the resources are there, the world will struggle to satisfy its thirst for oil, even at today’s prices. Consumers and producers are not at loggerheads. Producers have a sovereign right to determine what pace of development of their resources is in their national interest. Consumers respect this, though they also expect commitments to be met. But in any case, demand is booming in Middle East countries, too: they account for 20% of oil demand growth over the period to 2030. If prices rise still higher, they will drive even faster the search for non-oil solutions in OECD countries and oil-importing countries elsewhere. High oil prices do, over time, serve one purpose now almost unanimously accepted worldwide: to cut greenhouse-gas emissions to levels which will not cause irreparable damage to the world’s climate. Relative to WEO-2007, this year’s figures show lower global fossil-fuel energy use and greenhouse-gas emissions. But, despite the price changes, the reduction is nowhere near enough. Global energy-related greenhouse-gas emissions still increase by 45% by 2030.
We have two options. We can accept as broadly inevitable the outcome portrayed here in the first part of our analysis, which shows the destination of the course on which we are now set; and prepare ourselves to adapt to that uncertain even alarming, future. This path would lead to possible energy-related conflict and social disruption. Or we can plan and implement a new course, drawing on a united determination on the part Foreword
3
© OECD/IEA, 2008
This situation can be changed. Negotiators at Copenhagen in 2009 will seek to do that. Energy is a big part of the total climate change picture — over 60% of greenhouse-gas emissions come from energy production and use — but still only part. We show here how energy relates to the total picture and, in detail, how to go about making the energy future sustainable as part of a global climate solution. All nations would need to be involved, in a fair and proportionate manner.
of governments and action by committed citizens across the globe. The International Energy Agency sets out here what needs to be done to arrive at a supportable and sustainable future. The IEA will work with all nations, member countries and nonmember countries alike, to help effect the necessary changes. For those who wish to see, this WEO points the way. As last year, I pay tribute to Fatih Birol, who has conceived and directed this analysis, his team and the many others who have contributed. The calibre of their work matches, in depth, the profundity of the problems. Nobuo Tanaka Executive Director
4
World Energy Outlook 2008
© OECD/IEA, 2008
This publication has been produced under the authority of the Executive Director of the International Energy Agency. The views expressed do not necessarily reflect the views or policies of individual IEA member countries.
ACKNOWLEDGEMENTS This study was prepared by the Office of the Chief Economist (OCE) of the International Energy Agency (IEA) in co-operation with other offices of the Agency. The study was designed and directed by Fatih Birol, Chief Economist of the IEA. Trevor Morgan was in charge of co-ordinating the analysis on oil and gas supply; Laura Cozzi was responsible for the co-ordination of the analysis of climate-policy scenarios; Hideshi Emoto was in charge of modelling of energy demand, Maria Argiri of power generation analysis and Olivier Rech of oil-supply modelling. Teresa Malyshev co-ordinated the work on sub-Saharan Africa. Kamel Bennaceur, Raffaella Centurelli, Michael-Xiaobao Chen, Paul Dowling, Lorcan Lyons, Bertrand Magne, Chris Mullin, Uğur Öcal, Pawel Olejarnik, Fabien Roques, Olivier Sassi and Ralph Sims also authored different chapters of the book and were instrumental in delivering the study. Sandra Mooney provided essential support. For more information on the members of the OCE and their contribution to the World Energy Outlook, please visit www.worldenergyoutlook.org. Robert Priddle carried editorial responsibility. Our analysis of oil supply is based on a vast amount of field-by-field data that we compiled from a range of different sources. The primary source of this was IHS Energy, without whose assistance we would not have been able to carry out this work. We gratefully acknowledge their contribution. We are also indebted to the United Nations Framework Convention on Climate Change (UNFCCC) Secretariat for very helpful discussions when building our climate-change scenarios. Our strategic approach and analysis benefited greatly from the expertise of the World Energy Outlook’s two Advisory Panels, comprising prominent experts in the fields of oil and climate change. We are very grateful to our distinguished panel members for their guidance:
Advisory Panel on Oil Supply Prospects Guy Caruso (former Administrator of Energy Information Administration, US) Robert Fryklund (Vice President of Industry Relations, IHS) Leo Roodhart (President, Society of Petroleum Engineers, 2009) Ramzi Salman (former Deputy Secretary General, OPEC and former Advisor to the Minister of Energy, Qatar) Adnan Shihab-Eldin (former Secretary General, OPEC)
Yvo de Boer (Secretary General, UNFCCC) Rajendra K. Pachauri (Chairman, Intergovernmental Panel on Climate Change) Nicholas Stern (London School of Economics) David Victor (Stanford University) Dadi Zhou (former Director General, Energy Research Institute, National Development and Reform Commission, China) Acknowledgements
5
© OECD/IEA, 2008
Advisory Panel on Climate Change
A number of other international experts provided invaluable contributions throughout the preparation of this book: Paul Bailey (Department for Business, Enterprise and Regulatory Reform, UK), Paul Baruya (IEA Clean Coal Centre), Adam Chambers (IIASA), Donald Gautier (US Geological Survey), Jan-Hein Jesse (Shell), Bertrand Michel (Université Paris XI), Nebojsa Nakicenovic (Technical University of Vienna), Shaun Ragnauth (Environmental Protection Agency, US), Ivan Sandrea (StatoilHydro), and Detlef van Vuuren (Netherlands Environment Assessment Agency). The study benefited from input provided by IEA experts in different offices. Richard Bradley, Head of Energy Efficiency and Environment Division, made invaluable contributions to our climate-change analysis. The work also greatly benefited from the expertise of Nigel Jollands on cities, Ian Cronshaw on gas and Brian Ricketts on coal. Other IEA colleagues who provided input to different parts of the book include: Richard Baron, Aad van Bohemen, Amos Bromhead, Pierpaolo Cazzola, Ian Cronshaw, Muriel Custodio, Paolo Frankl, Lew Fulton, David Fyfe, Rebecca Gaghen, Jean-Yves Garnier, Dolf Gielen, Tim Gould, Hiroshi Hashimoto, Neil Hirst, Didier Houssin, Kierean Mcnamara, Andrea Nour, Cédric Philibert, Bertrand Sadin, Jonathan Sinton, Ulrik Stridbaek and Michael Taylor. Thanks also go to Marilyn Smith for proofreading the text. The work could not have been achieved without the substantial support and co-operation provided by many government bodies, international organisations and energy companies worldwide, notably: Deloitte and Touche, United Kingdom; the European Commission; Dutch Ministry of Economic Affairs; Foreign and Commonwealth Office, United Kingdom; Direction Générale de l’Energie et du Climat, France; IEA Clean Coal Centre; Korean Ministry of Commerce, Industry and Energy; Michelin, France; Norwegian Ministry of Foreign Affairs; Schlumberger Ltd; Shell; StatoilHydro; Sustainable Energy Ireland; Swiss Federal Department of the Environment, Transport, Energy and Communications; the Government of Denmark; Ministry of Economy, Trade and Industry, Japan; HM Treasury, United Kingdom; International Energy Forum, Saudi Arabia; the Renewable Energy and Energy Efficiency Partnership (REEEP), Austria; CIRED, Toyota Motor Corporation, Japan; US Department of Energy; US Environment; Protection Agency; US Geological Survey and US National Petroleum Council. Many international experts provided input, commented on the underlying analytical work, and reviewed early drafts of each chapter. Their comments and suggestions were of great value. They include:
Paul Alba
Consultant, France
Paul Bailey
Department for Business, Enterprise and Regulatory Reform, UK
Ashok Belani
Schlumberger, France
Johannes Benigni
JBC Energy, Austria
Jan Bygdevoll
Norwegian Petroleum Directorate
Alessandro Campo
Unicredit, Italy
6
World Energy Outlook 2008
© OECD/IEA, 2008
Oil supply prospects
Total, France
Bernhard Cramer
Federal Institute for Geosciences and Natural Resources, Germany
William Davie
Schlumberger, France
John Fitzgerald
The Economic and Social Research Institute, Ireland
Dario Garofalo
Ministry of Foreign Affairs, Italy
John Guy
National Petroleum Council, US
Gilbert Hamaide
GDF Suez, France
Sigurd Heiberg
StatoilHydro, Norway
Anouk Honoré
Oxford Institute for Energy Studies, UK
Peter Jackson
IHS Energy
Tor Kartevold
StatoilHydro, Norway
Jostein Dahl Karlsen
Ministry of Petroleum and Energy, Norway
Ryan Kauppila
Lehman Brothers, UK
Tim Klett
US Geological Survey
David Knapp
Energy Intelligence, US
Sally Kornfeld
Department of Energy, US
Ken Koyama
The Institute of Energy Economics, Japan
Alessandro Lanza
Eni, Italy
Colin Lyle
International Gas Union, UK
William Martin
Washington Policy and Analysis, US
Yves Mathieu
IFP, France
Kenneth McKellar
Deloitte and Touche, UK
Pedro Antonio Merino
Repsol, Spain
Nasarudin Md Idris
Petronas, Malaysia
Rodolphe Olard
ING, UK
Stephen Plumb
Chevron, US
Xavier Preel
Total, France
Dinko Raytchev
Directorate-General Energy and Transport, European Commission
Kenneth Rogoff
Harvard University, US
Rafael Sandrea
IPC Petroleum Consultants, US
Richard Schimpf
E.ON, Germany
Adam Sieminski
Deutsche Bank, US
Pierre Sigonney
Total, France
Bob Skinner
StatoilHydro, Norway
Jonathan Stern
Oxford Institute for Energy Studies, UK
Acknowledgements
7
© OECD/IEA, 2008
Joel Couse
Lynsey Tinios
Department for Business, Enterprise and Regulatory Reform, UK
Coby van der Linde
Clingendael International Energy Programme, Netherlands
Noé van Hulst
International Energy Forum, Saudi Arabia
Laszlo Varro
MOL Group, Hungary
Frank Verrastro
Center for Strategic and International Studies, US
Jun Arima
Ministry of Economy, Trade and Industry, Japan
Terry Barker
Cambridge Centre for Climate Policy Research, UK
Morgan Bazilian
Department of Communications, Energy and Natural Resources, Ireland
Christine Berg
Directorate-General Energy and Transport, European Commission
Geoffrey Blanford
Electric Power Research Institute, US
Jean-Paul Bouttes
EDF Energy, France
Peter Brun
Vestas, Denmark
Per Callesen
Danish Ministry of Finance
Christa Clapp
Environmental Protection Agency, US
Carmen Difiglio
Department of Energy, US
Bhava Dhungana
UNFCCC
Richard Folland
JP Morgan, UK
Peter Fraser
Ontario Energy Board, Canada
Christoph Frei
World Economic Forum, Switzerland
Dario Garofalo
Ministry of Foreign Affairs, Italy
Rainer Görgen
Federal Ministry of Economics and Technology, Germany
Michael Grubb
The Carbon Trust, UK
Reinhard Haas
Technical University of Vienna, Austria
Donald Hanson
Argonne National Laboratory, US
Colin Henderson
IEA Clean Coal Centre, UK
James Hewlett
Department of Energy, US
Takashi Hongo
Japan Bank for International Cooperation
John Houghton
Intergovernmental Panel on Climate Change
Jeff Huntington
European Environment Agency
Craig Jones
Department for Business, Enterprise and Regulatory Reform, UK
Julia King
Aston University, UK
8
World Energy Outlook 2008
© OECD/IEA, 2008
Climate change
Ger Klaassen
Directorate-General Environment, European Commission
Doug Koplow
Earthtrack, US
Takayuki Kusajima
Toyota, Japan
Kate Larsen
Department of State, US
Audrey Lee
Department of Energy, US
Joan MacNaughton
Alstom Power
Ritu Mathur
The Energy and Resources Institute, India
Neil McMurdo
HM Treasury, UK
Bert Metz
Netherlands Environmental Assessment Agency
Jan Mizgajski
Ministry of Economy, Poland
Arne Mogren
Vattenfall, Sweden
Helen Mountford
OECD Environment Directorate
Patrick Oliva
Michelin, France
Paul Oliver
World Energy Forum
Ayse Yasemin Orucu
Ministry of Energy and Natural Resources, Turkey
John Paffenbarger
Constellation Energy, US
Binu Parthan
Renewable Energy and Energy Efficiency Partnership, Austria
Gustav Resch
Vienna University of Technology, Austria
Hans-Holger Rogner
International Atomic Energy Agency
Leo Schrattenholzer
IRM, Austria
PR Shukla
Indian Institute of Management
Bjorn Stigson
World Business Council for Sustainable Development
Sven Teske
Greenpeace
Fridtjof Unander
Enova, Norway
Tom Van Ierland
Directorate-General Environment, European Commission
Henning Wüster
UNFCCC
Arthouros Zervos
Global Wind Energy Council
Edward Caldwell
ERC Consultants, Canada
Heleen de Coninck
Energy Research Centre of the Netherlands
Karim Dahou
OECD Directorate for Financial and Enterprise Affairs
Stanislas Drochon
PFC Energy, France
Jean-Pierre Favennec
IFP, France
Stephen Gitonga
UN Development Programme
Beejaye Kokil
African Development Bank
Michael Levitsky
World Bank
Acknowledgements
9
© OECD/IEA, 2008
Sub-Saharan Africa
Vijay Modi Petter Nore Yinke Omorogbe Francisco Paris George Person
Columbia University, US Norwegian Agency for Development Cooperation University of Ibadan, Nigeria Extractive Industries Transparency Initiative, Norway Department of Energy, US
Jürgen Reitmaier Kenneth Ruffing Connie Smyser Luigi Tessiore Noé van Hulst
Extractive Industries Transparency Initiative, Norway OECD Development Centre Smyser Associates, US UN Development Programme, Senegal International Energy Forum, Saudi Arabia
Hans Terje Ylvisåker
Norwegian Embassy, Mozambique
Energy use in cities Shobhakar Dhakal Leticia Guimaraes Stephen Hammer Stephen Kenihan Lily Parshall Matthias Ruth Niels Schulz Wayne Wescott
Global Carbon Project, Japan University of Maryland, US Columbia University, US ICLEI – Oceania, Australia Columbia University, US University of Maryland, US Imperial College London, UK ICLEI – Oceania, Australia
10
World Energy Outlook 2008
© OECD/IEA, 2008
The individuals and organisations that contributed to this study are not responsible for any opinions or judgements contained in this study. All errors and omissions are solely the responsibility of the IEA.
Comments and questions are welcome and should be addressed to: Dr. Fatih Birol Chief Economist Director, Office of the Chief Economist International Energy Agency 9, rue de la Fédération 75739 Paris Cedex 15 France (33-1) 4057 6670 (33-1) 4057 6659
[email protected] 11
© OECD/IEA, 2008
Telephone: Fax: Email:
T A B L E
PART A GLOBAL ENERGY TRENDS TO 2030
O F
PART C THE ROLE OF ENERGY IN CLIMATE POLICY
ANNEXES
© OECD/IEA, 2008
C O N T E N T S
PART B OIL AND GAS PRODUCTION PROSPECTS
CONTEXT AND PRINCIPAL ASSUMPTIONS
FOREWORD
1
GLOBAL ENERGY TRENDS
2
OIL MARKET OUTLOOK
3
NATURAL GAS MARKET OUTLOOK
4
COAL MARKET OUTLOOK
5
POWER SECTOR OUTLOOK
6
RENEWABLE ENERGY OUTLOOK
7
ENERGY USE IN CITIES
8
TURNING OIL RESOURCES INTO RESERVES
9 10
PROSPECTS FOR OIL PRODUCTION
11
NATURAL GAS RESOURCES AND PRODUCTION PROSPECTS
12
UPSTREAM INVESTMENT PROSPECTS
13
THE STRUCTURE OF THE UPSTREAM INDUSTRY
14
PROSPECTS IN OIL- AND GAS-EXPORTING SUB-SAHARAN AFRICAN COUNTRIES
15
IMPLICATIONS OF THE REFERENCE SCENARIO FOR THE GLOBAL CLIMATE
16
THE POST-2012 CLIMATE POLICY FRAMEWORK
17
CLIMATE POLICY SCENARIOS
18
IMPLICATIONS FOR ENERGY INVESTMENT
ANNEXES
© OECD/IEA, 2008
FIELD-BY-FIELD ANALYSIS OF OIL PRODUCTION
19
Foreword Acknowledgements List of Figures List of Tables List of Boxes List of Spotlights Executive Summary Introduction
3 5 20 28 32 34 37 51
Part A: Global energy trends to 2030
56
2
3
14
Context and principal assumptions Highlights
59 59
Government policies and measures
60
Population Economic growth Energy prices Oil prices Natural gas prices Steam coal prices Technology
63 64 68 68 72 73 73
Global energy trends Highlights Demand Primary energy mix Regional trends Sectoral Trends Energy production and trade Resources and production prospects Inter-regional trade Energy investment Trends by region and energy source
77 77 78 78 80 83 85 85 87 88 88
Oil market outlook Highlights Demand Trends in primary oil demand Regional trends Sectoral trends Implications for spending on oil Production Inter-regional trade
91 91 92 92 97 98 100 102 105 World Energy Outlook 2008
© OECD/IEA, 2008
1
Natural gas market outlook Highlights Demand Regional trends Sectoral trends Production Inter-regional trade
109 109 110 112 113 115 119
5
Coal market outlook Highlights Demand Reserves and production Inter-regional trade Coal prices and supply costs Coal investment
123 123 124 127 131 134 136
6
Power sector outlook Highlights Electricity demand Electricity supply Outlook for electricity generation Trends in coal-fired generation Trends in gas-fired generation Trends in oil-fired generation Trends in nuclear power Trends in renewable energy Trends in CO2 capture and storage New capacity and investment needs in infrastructure Electricity generating costs Trends in construction costs Cost and efficiency assumptions Electricity prices
139 139 140 142 142 144 146 147 147 149 149 151 152 152 153 156
Renewable energy outlook Highlights Global trends in the use of renewable energy Renewables for electricity Hydropower Wind power Solar photovoltaics Concentrating solar thermal power Geothermal power Tide and wave power Biomass for electricity Biofuels
159 159 160 162 165 166 168 170 170 171 171 171
7
Table of contents
15
© OECD/IEA, 2008
4
176 177 178
Energy use in cities Highlights Why focus on cities? Current and projected energy use in cities United States Background Reference Scenario projections European Union Background Reference Scenario projections Australasia Background Reference Scenario projections China Background Reference Scenario projections
179 179 180 182 184 184 185 187 187 188 189 189 190 191 191 192
Part B: Oil and gas production prospects
194
Turning oil resources into reserves Highlights
197 197
9
Classifying hydrocarbon resources Latest estimates of conventional oil reserves and resources Proven reserves Ultimately recoverable resources
198 201 201 204
Reserves growth and enhanced oil recovery Carbon-dioxide enhanced oil recovery (CO2-EOR) Non-conventional oil resources
209 213 215
Extra-heavy oil and oil sands Oil shales: a technological milestone in sight?
10
16
215 217
Long-term oil-supply cost curve
217
Field-by-field analysis of oil production Highlights
221 221
Understanding production patterns and trends
222
The importance of size Regional differences Oilfield production profiles and characteristics Standard production profiles Focus on giant fields Changes in production profiles over time
225 227 229 229 232 233 World Energy Outlook 2008
© OECD/IEA, 2008
8
Renewables for heat Traditional biomass Investment in renewable energy
Measuring observed production decline rates
233
Approach and definitions 233 Results of the analysis 236 The impact of field age and maturity 240 Trends in observed decline rates 241 Deriving an estimate of the average global observed decline rate 242 Estimating historical trends Long-term prospects for natural decline rates
11
12
13
243 243 247
Prospects for oil production Highlights Global oil production trends Summary of projections in the Reference Scenario Crude oil output at existing fields Contribution of new fields to crude oil production Enhanced oil recovery (EOR) Natural gas liquids (NGLs) Non-conventional oil Crude oil quality Outlook by country and region Non-OPEC production OPEC production Sensitivity of oil output to decline rates
249 249 250 250 255 256 260 261 262 265 266 266 272 276
Natural gas resources and production prospects Highlights Gas resources and reserves Conventional gas Non-conventional gas Gas production prospects Regional trends OECD North America OECD Europe OECD Oceania Eastern Europe/Eurasia Non-OECD Asia Middle East Africa Latin America
279 279 280 280 284 289 291 291 293 294 294 297 298 300 302
Upstream investment prospects Highlights Recent investment trends and near-term outlook New upstream projects Capital-cost trends
303 303 304 310 311
Table of contents
17
© OECD/IEA, 2008
Trends in natural decline rates
15
320 322 323 325 325 328 329 329
The structure of the upstream industry Highlights The emergence of a new world order for oil The resurgence of the national oil companies International oil companies in profitable retreat Implications for future investment and supply Towards a more efficient industry Strengthening strategic partnerships
333 333 334 336 343 348 350 351
Prospects in oil- and gas-exporting sub-Saharan African countries Highlights Overview Outlook for oil and gas production, exports and government revenues Resources and reserves Oil and gas production and exports Oil refining Oil and gas export revenues Household energy access in sub-Saharan African countries Cooking practices Electricity access Projected trends in access to modern energy services Quantifying the costs involved in expanding access Managing revenues from oil and gas
355 355 356 358 358 359 363 364 364 365 366 369 370 374
Part C: The role of energy in climate policy 16
18
378
Implications of the Reference Scenario for the global climate Highlights Projected trends in energy-related CO2 emissions Overview Regional trends Trends per capita and per unit of GDP Sectoral trends Projected trends in overall greenhouse-gas emissions Long-term greenhouse-gas concentrations and average global temperature The impact of global warming on the energy sector Hydropower Other renewable energy systems
381 381 382 382 384 388 391 397 401 402 402 403
World Energy Outlook 2008
© OECD/IEA, 2008
14
Trends in upstream activity Implications for oil-production capacity Outlook for investment to 2030 Potential barriers to upstream investment Depletion policies of resource-rich countries Profitable opportunities for international companies to invest Political constraints Availability of people and equipment
Thermal and nuclear power Local and regional air pollution Projected trends in local and regional air pollution
18
19
The post-2012 climate policy framework 407 Highlights 407 Co-ordinated global action to address climate change 408 Defining the long-term global climate objective 410 Environmental and societal effects of different stabilisation levels 410 Practical considerations in reducing emissions 411 Climate scenarios modelled in WEO-2008 413 Participation 416 Principles for allocating responsibilities 419 Policy mechanisms 421 Cap-and-trade systems 423 Sectoral agreements 427 National policies and measures 432 Climate policy scenarios Highlights Methodology and assumptions Modelling approach Economic growth Energy prices Carbon prices Overview of the results of the climate scenarios Primary energy demand Energy-related CO2 emissions Emissions of gases other than energy-related CO2 The 550 Policy Scenario: results by sector Power generation Industry Transport Buildings and other sectors The 450 Policy Scenario: results by sector Power generation Industry Transport Buildings and other sectors Co-benefits in the 550 and 450 Policy Scenarios Energy security Local and regional air pollution
435 435 436 436 437 437 439 439 439 442 447 450 450 457 461 464 467 467 472 473 474 475 475 476
Implications for energy investment Highlights The investment challenge of transforming energy Investment implications of the 550 Policy Scenario
479 479 480 480
Table of contents
19
© OECD/IEA, 2008
17
404 404 404
ANNEXES Annex A. Annex B. Annex C. Annex D.
Regional implications of the 550 Policy Scenario Implications of the 550 Policy Scenario for the electricity sector Implications of the 550 Policy Scenario for the transport sector Implications of the 550 Policy Scenario for buildings and industry Costs and benefits of the 550 Policy Scenario Investment implications of the 450 Policy Scenario Replacement of capital stock in the power sector Technology diffusion and interaction with carbon markets Carbon markets Technology diffusion Implications for policy
481 483 485 487 487 488 492 495 495 499 501
Tables for Reference Scenario projections Abbreviations, definitions and conversion factors Acronyms References
503 505 541 551 555
List of figures Part A: Global energy trends to 2030
Chapter 2. Global energy trends 2.1 World primary energy demand by fuel in the Reference Scenario 2.2 World primary energy demand by region in the Reference Scenario 2.3 Incremental primary energy demand by fuel in the Reference Scenario, 2006-2030 2.4 Per-capita primary energy demand by region, 2030 2.5 Incremental world fossil-fuel production in the Reference Scenario 2.6 Cumulative investment in energy-supply infrastructure in the Reference Scenario, 2007-2030 Chapter 3. Oil market outlook 3.1 Change in world primary oil demand and real GDP growth 3.2 Oil intensity by region in the Reference Scenario 3.3 Global oil demand and the oil price in the three oil shocks 3.4 Average IEA crude oil import price in different currencies 3.5 Change in primary oil demand by region in the Reference Scenario, 2007-2030 20
62 63 67 69 72 75 80 81 82 83 87 89 93 94 95 95 97
World Energy Outlook 2008
© OECD/IEA, 2008
Chapter 1. Context and principal assumptions 1.1 Energy subsidies in non-OECD countries, 2007 1.2 Population by major region 1.3 Rate of growth of per-capita income by region 1.4 Average IEA crude oil import price 1.5 Assumed natural gas and coal prices relative to crude oil 1.6 Typical lifetimes of energy-related capital stock
3.9 3.10 3.11
Incremental oil demand by sector in the Reference Scenario, 2006-2030 Light-duty vehicle stock by region in the Reference Scenario Share of oil spending in real GDP at market exchange rates in the Reference Scenario World oil production by source in the Reference Scenario Oil-import dependence by major importing region in the Reference Scenario Oil export flows from the Middle East
Chapter 4. Natural gas market outlook 4.1 Change in gas intensity by region in the Reference Scenario 4.2 Increase in primary demand for natural gas by region in the Reference Scenario 4.3 Increase in world primary natural gas demand by sector in the Reference Scenario, 2006-2030 4.4 World primary natural gas demand by sector in the Reference Scenario 4.5 Natural gas production by region in the Reference Scenario 4.6 Main net inter-regional natural gas trade flows in the Reference Scenario, 2006 and 2030 4.7 World inter-regional natural gas trade by type in the Reference Scenario 4.8 Inter-regional exports of LNG by source in the Reference Scenario Chapter 5. Coal market outlook 5.1 Incremental world primary energy demand by fuel, 2000-2006 5.2 Sectoral shares in coal demand by region in the Reference Scenario 5.3 World energy-related CO2 emissions by fuel in the Reference Scenario 5.4 Proven coal reserves in leading producing countries, 2005 5.5 World coal production by type in the Reference Scenario 5.6 Net inter-regional hard coal trade in the Reference Scenario 5.7 Index of selected fuel prices 5.8 Coal supply cash-cost curve for internationally traded steam coal for 2007 and selected average FOB prices for 2007 and first-half 2008 5.9 Structure of coal-supply costs for major exporting countries 5.10 Cumulative investment in coal-supply infrastructure by region in the Reference Scenario, 2007-2030 Chapter 6. Power sector outlook 6.1 Electricity demand growth rates by region in the Reference Scenario 6.2 Per-capita electricity demand by selected region in the Reference Scenario 6.3 World electricity generation by fuel in the Reference Scenario 6.4 Power-generation capacity under construction worldwide 6.5 Coal-fired capacity under construction in OECD countries 6.6 Efficiency improvements in coal-fired generation in the Reference Scenario 6.7 Commodity price indices 6.8 Electricity generating costs in selected regions 6.9 Indices of real end-use fuel and electricity prices in OECD 6.10 European wholesale prices of electricity and EU emission allowances Table of contents
98 100 102 104 105 107 112 112 113 114 116 120 121 121 124 126 127 128 129 133 135 136 137 137 141 142 143 144 145 146 152 154 156 157 21
© OECD/IEA, 2008
3.6 3.7 3.8
Chapter 7. Renewable energy outlook 7.1 Contribution of biomass to world primary energy demand, 2006 7.2 Share of total electricity generation from renewables by region in the Reference Scenario 7.3 Increase in world electricity generation from renewables in the Reference Scenario 7.4 Increase in OECD electricity generation by energy source in the Reference Scenario 7.5 Projected generating costs of renewable energy technologies in the Reference Scenario 7.6 Projected investment costs of renewable energy technologies in the Reference Scenario 7.7 Hydropower capacity under construction by region 7.8 World biofuels consumption by type in the Reference Scenario 7.9 World investment in new power-generation plants by fuel in the Reference Scenario Chapter 8. Energy use in cities 8.1 Regional trends in urbanisation 8.2 World and city primary energy consumption in the Reference Scenario 8.3 US energy consumption in cities by state, 2006 8.4 Per-capita energy demand in the European Union and EU cities in the Reference Scenario 8.5 Per-capita energy demand and gross regional product in selected Chinese cities, 2006
161 162 163 163 164 165 166 172 178 180 184 186 189 192
Part B: Oil and gas production prospects 199 202 203 203 205 208 209 210 215 218
Chapter 10. Field-by-field analysis of oil production 10.1 World crude oil production from super-giant and giant fields by field vintage 227 10.2 Crude oil production by region and size of field, 2007 228 22
World Energy Outlook 2008
© OECD/IEA, 2008
Chapter 9. Turning oil resources into reserves 9.1 Hydrocarbon resource classification 9.2 Estimated remaining world oil reserves, end-2007 9.3 Proven remaining oil reserves by region, 1980-2007 9.4 Oil discoveries and production, 1960-2006 9.5 New field discoveries and reserves growth in 1996-2003 compared with USGS 2000 assessments 9.6 Global offshore oil discoveries to end-2006 versus exploration wells 9.7 Maximum water depth of offshore exploration and production wells worldwide 9.8 A case study of oil reserves growth: the impact of technology on oil production from the Weyburn field in Canada 9.9 Potential additional recoverable oil resources using CO2-EOR by region 9.10 Long-term oil-supply cost curve
10.6 10.7 10.8 10.9 10.10 10.11 10.12
Standard oilfield-production profiles by category of field Standard oilfield-production profiles of giant fields Selected production profiles of recently developed medium-sized onshore fields compared with the standardised profile Production-weighted average post-peak and post-plateau observed decline rates by field size Production-weighted average post-peak observed decline rates by type of producer and year of first production Year-on-year change in the production-weighted average production from post-peak fields Methodology for estimating natural decline rates Indicative natural decline rates by region Natural decline rates and reserves-to-production ratios by region, 2007 Projected change in natural decline rates and reserves-to-production ratios by region, 2007 to 2030
232 233 234 239 240 242 244 245 247 248
Chapter 11. Prospects for oil production 11.1 World oil production by source in the Reference Scenario 11.2 Architecture of the Oil Supply Model 11.3 Crude oil production from existing fields in OPEC and non-OPEC countries in the Reference Scenario 11.4 Crude oil production decline of existing fields by region in the Reference Scenario, 2007-2030 11.5 World crude oil production from new fields in the Reference Scenario 11.6 Crude oil production from yet-to-be-developed fields in OPEC and non-OPEC countries by location in the Reference Scenario 11.7 Conventional proven and probable crude oil reserves in yet-to-be-developed fields by region, end-2007 11.8 Number of yet-to-be-developed oilfields by region and location, end-2007 11.9 World crude oil production of yet-to-be-found oilfields in OPEC and non-OPEC countries by location in the Reference Scenario 11.10 Enhanced oil recovery by country in the Reference Scenario 11.11 World natural gas liquids production by OPEC and non-OPEC countries in the Reference Scenario 11.12 World non-conventional oil production by type in the Reference Scenario 11.13 Crude oil quality worldwide in the Reference Scenario 11.14 Change in oil production by country/region, 2007-2030 11.15 Non-OPEC oil production by type in the Reference Scenario 11.16 OPEC oil production by type in the Reference Scenario
262 265 266 268 273
Chapter 12. Natural gas resources and production prospects 12.1 Proven reserves of natural gas 12.2 Regional share in natural gas production and proven reserves, 2007 12.3 Natural gas discoveries and cumulative production, 1960-2006
280 281 281
Table of contents
250 254 256 256 257 258 258 259 260 261 261
23
© OECD/IEA, 2008
10.3 10.4 10.5
Tight gas drilling in the United States Location of gas-hydrate resources Change in natural gas production by country/region, 2007-2030 US natural gas balance Natural gas production in Eastern Europe/Eurasia
Chapter 13. Upstream investment prospects 13.1 World investment in oil and gas exploration and production 13.2 Upstream investment of surveyed companies by type of company 13.3 Upstream oil and gas investment of surveyed companies by activity 13.4 Investment in oil and gas exploration of surveyed companies 13.5 Investment in new upstream oil and gas projects, 2008-2015 13.6 Finding and development costs for US FRS companies 13.7 Worldwide upstream drilling cost indices 13.8 Day-rates for drilling rigs 13.9 Drilling rig day-rates and the average IEA crude oil import price, 1995-2007 13.10 Average capital cost of upstream projects under development 13.11 IEA Upstream Investment Cost Index by activity 13.12 Offshore rigs under construction and effective utilisation rates 13.13 Year-on-year change in world upstream investment and dollar-cost inflation rate 13.14 Indicators of upstream activity 13.15 Gross peak-oil production capacity additions from current projects by region, 2008-2015 13.16 Cumulative upstream investment by region in the Reference Scenario, 2007-2030 13.17 OPEC oil and gas export revenues in the Reference Scenario 13.18 OPEC countries’ external account price thresholds to balance the external account 13.19 Demand for and supply of mid-career petrotechnical professionals aged 30 to 39 in North America and Europe Chapter 14. The structure of the upstream industry 14.1 World oil and gas reserves and production by company, 2007 14.2 Foreign company access to proven oil reserves, end-2007 14.3 Share of national oil and gas companies’ upstream production from overseas assets, 2007 14.4 Year-on-year increase in oil production of surveyed companies by type of company 14.5 Net income of the super-majors versus the average IEA crude oil import price, 2000-2007 14.6 International oil company outlays from operating cash-flow 14.7 International oil company oil and gas reserves replacement ratio 14.8 Return on average capital employed of the super-majors 24
285 289 291 293 296
304 306 308 309 310 311 315 316 316 317 318 319 321 321 322 324 326 327 330
335 336 342 343 344 345 346 347
World Energy Outlook 2008
© OECD/IEA, 2008
12.4 12.5 12.6 12.7 12.8
14.9 14.10
World oil and gas production by type of company in the Reference Scenario Revenue and hydrocarbon production per employee by type of company, 2007
349 350
Chapter 15. Prospects in oil- and gas-exporting sub-Saharan African countries 15.1 Share of oil and gas in total exports and of oil and gas revenues in government revenues in assessed sub-Saharan African countries, 2006 357 15.2 Proven oil and gas reserves in assessed sub-Saharan African countries, end-2007 359 15.3 Oil and gas production in assessed sub-Saharan African countries 360 15.4 Gas flaring in assessed sub-Saharan African countries 362 15.5 Oil and gas supply and infrastructure in sub-Saharan Africa 363 15.6 Cumulative government oil and gas revenues in assessed sub-Saharan African countries, 2006-2030 364 15.7 Regional power pools in Africa 368 15.8 Total additional cost of universal access to electricity and clean cooking stoves in assessed sub-Saharan African countries 371 Part C: The role of energy in climate policy
Table of contents
382 384 385 386 388 389 391 392 392 394 395 396 398 25
© OECD/IEA, 2008
Chapter 16. Implications of the Reference Scenario for the global climate 16.1 Energy-related CO2 emissions in the Reference Scenario by fuel and region 16.2 Average annual growth in world primary energy demand and energy related CO2 emissions in the Reference Scenario 16.3 Average annual increase in energy-related CO2 emissions by region in the Reference Scenario 16.4 Incremental energy-related CO2 emissions by country and region, 2006-2030 16.5 Ten largest inter-regional flows of energy-related CO2 emissions embedded in exports of goods and services, 2006 16.6 Average annual rate of change in energy-related CO2 emissions per unit of GDP by region, 2006-2030 16.7 Energy-related CO2 emissions in cities by region in the Reference Scenario 16.8 Energy-related CO2 emissions from power plants by fuel and region in the Reference Scenario 16.9 CO2 intensity of power generation by region in the Reference Scenario 16.10 Change in transport sector CO2 emissions by mode and region in the Reference Scenario, 2006-2030 16.11 Industry sector energy-related CO2 emissions by region in the Reference Scenario 16.12 Increase in direct and indirect energy-related CO2 emissions by region in the residential, services and agriculture sectors, 2006-2030 16.13 World anthropogenic greenhouse-gas emissions by source, 2005
16.15 16.16
World anthropogenic greenhouse-gas emissions by source in the Reference Scenario, 2005-2030 Total greenhouse-gas emissions other than energy-related CO2 by region in the Reference Scenario Energy-related CO2 emissions in the Reference Scenario to 2030 compared with other published scenarios to 2100
Chapter 17. The post-2012 climate policy framework 17.1 Potential effects of stabilisation of atmospheric concentrations of greenhouse gases at different levels 17.2 Energy-related CO2 emissions from existing and future power plants in the Reference Scenario 17.3 Greenhouse-gas concentration trajectories by scenario 17.4 Energy-related CO2 emissions in the 550 Policy Scenario and 450 Policy Scenario 17.5 Regional shares of world energy-related CO2 emissions in the Reference Scenario for different levels of participation, 2020 17.6 Reduction in world energy-related CO2 emissions in the 450 and 550 Policy Scenarios compared with total OECD emissions in the Reference Scenario in 2030 17.7 Hybrid policy framework assumed in the 450 and 550 Policy Scenarios 17.8 Suitability of sectors to sectoral agreements Chapter 18. Climate policy scenarios 18.1 Change in electricity prices in OECD+ countries in the 550 Policy Scenario compared with the Reference Scenario 18.2 Change in primary energy demand in the 550 Policy Scenario relative to the Reference Scenario by fuel and region, 2030 18.3 Energy-related CO2 emissions reduction by region in the 550 and 450 Policy Scenarios relative to the Reference Scenario 18.4 Energy-related CO2 emissions by source in the 550 and 450 Policy Scenarios relative to the Reference Scenario 18.5 Global emissions of all greenhouse gases in the Reference, 550 and 450 Policy Scenarios 18.6 CO2 emissions per MWh of electricity generated in the 550 Policy Scenario compared with the Reference Scenario 18.7 Fuel shares in world electricity generation in the Reference and 550 Policy Scenarios 18.8 Change in coal- and gas-fired generation by region in the 550 Policy Scenario 18.9 World non-hydropower renewable electricity generation by source in the 550 Policy Scenario 18.10 Share of nuclear power in electricity generation by region in the 550 Policy Scenario 18.11 Energy-related CO2 emissions from industry by region in the Reference and 550 Policy Scenarios 26
400 400 402
412 413 414 415 417 418 422 430
438 440 443 446 449 452 453 454 455 456 458
World Energy Outlook 2008
© OECD/IEA, 2008
16.14
18.13 18.14 18.15 18.16 18.17 18.18 18.19 18.20 18.21
CO2 intensity improvement in the iron and steel sector and the cement sector by region in the Reference and 550 Policy Scenarios Energy-related CO2 emissions from the iron and steel and non-metallic minerals sectors in the Reference Scenario, 2030 CO2 intensity standards of light-duty vehicles by region in the 550 Policy Scenario Change in energy-related CO2 emissions in the buildings and other sectors in the Reference and 550 Policy Scenarios, 2006-2030 CO2 emissions reduction from the power sector in the 550 and 450 Policy Scenarios relative to the Reference Scenario World electriticity generation fuel mix in the 450 Policy Scenario relative to the 550 Policy Scenario Electricity generation by fuel and by regional grouping in the 450 Policy Scenario, 2030 Shares in world electricity generation by type of coal plant in the Reference, 550 and 450 Policy Scenarios OECD+ net oil and gas imports in the Reference and 550 Policy Scenarios, 2030 Percentage reduction in NOx and SO2 emissions by region in the 550 Policy Scenario relative to the Reference Scenario
Chapter 19. Implications for energy investment 19.1 Change in total world energy-related investments by sector in the 550 Policy Scenario relative to the Reference Scenario, 2010-2030 19.2 Additional investment in power plants and energy efficiency by region in the 550 Policy Scenario relative to the Reference Scenario, 2010-2030 19.3 Investment in the power sector in Other Major Economies and Other Countries in the 550 Policy Scenario and Reference Scenario, 2010-2030 19.4 Additional investment per capita by private consumers in the 550 Policy Scenario relative to the Reference Scenario, 2010-2030 19.5 Electricity generating costs in Europe and the United States assuming different carbon prices in the 550 Policy Scenario, 2030 19.6 Average light-duty vehicle fleet efficiency by region, 2006 19.7 Fuel cost savings by region in the 550 Policy Scenario relative to the Reference Scenario, 2010-2030 19.8 Change in power plant and energy efficiency investments in the Policy Scenarios relative to the Reference Scenario 19.9 Additional energy-related investment by region in the 450 Policy Scenario relative to the Reference Scenario, 2010-2030 19.10 Additional energy-related investment per private consumer in the 450 Policy Scenario relative to the Reference Scenario, 2021-2030 19.11 Age distribution of power plants in the OECD+ region, 2006 Table of contents
460 461 463 465 468 469 470 471 475 477
481
482
482 483 484 486 488 489 490 490 492 27
© OECD/IEA, 2008
18.12
19.12 19.13
Global carbon market trading volumes and values, 2006 and 2007 Revenues from full auctioning of CO2 allowances compared with additional investment in the 450 and 550 Policy Scenarios relative to the Reference Scenario, 2013-2030
496 499
List of tables Part A: Global energy trends to 2030
Chapter 2. Global energy trends 2.1 World primary energy demand by fuel in the Reference Scenario 2.2 World primary energy demand by region in the Reference Scenario 2.3 World final energy consumption by sector in the Reference Scenario 2.4 Cumulative investment in energy-supply infrastructure in the Reference Scenario, 2007-2030
60 64 66 68 78 81 84 88
Chapter 3. Oil market outlook 3.1 World primary oil demand in the Reference Scenario 3.2 Share of transport in primary oil demand by region in the Reference Scenario 3.3 World oil supply in the Reference Scenario 3.4 Net inter-regional oil trade in the Reference Scenario
99 103 106
Chapter 4. Natural gas market outlook 4.1 World primary demand for natural gas in the Reference Scenario 4.2 World natural gas production in the Reference Scenario 4.3 Net inter-regional natural gas trade in the Reference Scenario 4.4 Natural gas liquefaction capacity
110 115 118 122
Chapter 5. Coal market outlook 5.1 World primary coal demand in the Reference Scenario 5.2 World coal production in the Reference Scenario 5.3 Net inter-regional hard coal trade in the Reference Scenario
125 129 132
Chapter 6. Power sector outlook 6.1 World final electricity consumption by region in the Reference Scenario 6.2 Nuclear power plants under construction as of end-August 2008
140 148
28
93
World Energy Outlook 2008
© OECD/IEA, 2008
Chapter 1. Context and principal assumptions 1.1 Major new energy-related policy initiatives adopted between mid-2007 and mid-2008 1.2 Population growth by region 1.3 Real GDP growth by region 1.4 Fossil-fuel price assumptions
6.3 6.4 6.5
Recent CCS proposals and developments Projected capacity additions and investment needs in power infrastructure in the Reference Scenario Factors influencing power plant pricing
Chapter 7. Renewable energy outlook 7.1 Top-ten wind power markets, 2006 7.2 Final consumption of biofuels by region in the Reference Scenario Chapter 8. Energy use in cities 8.1 Overview of city energy use and urbanisation rate in regions and countries analysed in-depth, 2006 8.2 World energy demand in cities by fuel in the Reference Scenario 8.3 US energy demand in cities by fuel in the Reference Scenario 8.4 European Union energy demand in cities by fuel in the Reference Scenario 8.5 Australasian energy demand in cities by fuel in the Reference Scenario 8.6 Chinese energy demand in cities by fuel in the Reference Scenario
150 151 153 166 172
182 183 187 189 191 193
Part B: Oil and gas production prospects
Chapter 10. Field-by-field analysis of oil production 10.1 The world’s 20 biggest oilfields by production 10.2 World crude oil production by output and age of field 10.3 Geographical distribution of the world’s super-giant and giant oilfields 10.4 Average depletion factor of producing fields by size, 2007 10.5 Initial reserves of oilfield dataset for production profiling 10.6 Production characteristics of sample oilfield dataset for production profiling 10.7 Number of oilfields in dataset for decline rate calculations 10.8 Production-weighted average observed decline rates by size and type of field 10.9 Production-weighted average annual observed decline rates by region 10.10 Production-weighted average post-peak observed decline rates by vintage Table of contents
205 213 214 216 225 226 227 229 230 231 236 238 239 240 29
© OECD/IEA, 2008
Chapter 9. Turning oil resources into reserves 9.1 World ultimately recoverable conventional oil and NGL resources, end-2007 9.2 Estimated 2007 incremental production from EOR projects by technology 9.3 Pre-screening criteria for EOR 9.4 Extra-heavy oil and oil sands resources
10.12 10.13
Production-weighted average annual observed decline rates by decline phase 241 Estimated production-weighted average annual observed post-peak decline rates for all fields worldwide by region 243 Average year-on-year decline rates for selected North Sea oilfields 246
Chapter 11. Prospects for oil production 11.1 World oil production and supply in the Reference Scenario 11.2 Average size of yet-to-be-developed oilfields by region, end-2007 11.3 Non-OPEC oil production in the Reference Scenario 11.4 OPEC oil production in the Reference Scenario 11.5 World cumulative investment in the upstream oil sector under different decline rate assumptions for existing fields, 2007-2030 Chapter 12. Natural gas resources and production prospects 12.1 Remaining proven reserves and mean estimates of ultimately recoverable conventional resources of natural gas, end-2007 12.2 World proven sour gas reserves, end-2006 12.3 Enhanced coalbed methane CO2 sequestration potential by country/region 12.4 Natural gas production in the Reference Scenario 12.5 Qatari LNG projects Chapter 13. Upstream investment prospects 13.1 Companies in investment survey, 2007 13.2 New upstream oil and gas projects, 2008-2015 13.3 Cumulative upstream investment by region in the Reference Scenario, 2007-2030 Chapter 14. The structure of the upstream industry 14.1 Key indicators of leading oil and gas companies, 2007
251 259 267 272 277
282 283 287 290 299
307 312 324
338
Chapter 15. Prospects in oil- and gas-exporting sub-Saharan African countries 15.1 Production and reserves in assessed sub-Saharan African countries 356 15.2 Number of people without access to electricity and relying on fuelwood and charcoal for cooking in assessed sub-Saharan African countries 358 15.3 Oil production maturity of assessed sub-Saharan African countries, 2007 360 15.4 Annual premature death and disability associated with indoor air pollution in assessed sub-Saharan African countries 366 15.5 Access to electricity and reliance on fuelwood and charcoal in assessed sub-Saharan African countries 370 30
World Energy Outlook 2008
© OECD/IEA, 2008
10.11
15.6
Cumulative cost of providing universal access to modern energy in assessed sub-Saharan African countries, 2006-2030
374
Part C: The role of energy in climate policy
Chapter 17. The post-2012 climate policy framework 17.1 The world’s five biggest emitters of energy-related CO2 17.2 Change in 2020 of regional energy-related CO2 emissions, under different allocation mechanisms, to achieve a 10% reduction in global emissions relative to the Reference Scenario 17.3 Existing, announced and proposed emissions-trading systems 17.4 Examples of city climate-change policy targets Chapter 18. Climate policy scenarios 18.1 Macroeconomic cost of mitigation 18.2 World primary energy demand in the 550 Policy Scenario 18.3 World primary energy demand in the 450 Policy Scenario 18.4 Energy-related CO2 emissions by sector and fuel in the 550 and 450 Policy Scenarios 18.5 Major initiatives to reduce energy-related CO2 emissions in Other Major Economies 18.6 Electricity generation and related CO2 emissions in the 550 Policy Scenario compared with the Reference Scenario 18.7 Capacity additions in the 550 Policy Scenario 18.8 Energy demand and energy-related CO2 emissions in industry in the 550 Policy Scenario 18.9 Energy demand and energy-related CO2 emissions in transport in the 550 Policy Scenario 18.10 Energy demand and energy-related CO2 emissions by fuel and region in the buildings and other sectors in the 550 Policy Scenario 18.11 Power generation and energy-related CO2 emissions in the 450 Policy Scenario, 2030 18.12 Capacity additions in the 450 Policy Scenario relative to the 550 Policy Scenario, 2020-2030 Table of contents
383 385 389 393 405
417 420 425 433
437 439 441 444 445 451 456 458 462 467 468 471 31
© OECD/IEA, 2008
Chapter 16. Implications of the Reference Scenario for the global climate 16.1 Impact of retail-price increases on primary energy demand and energy-related CO2 emissions by region, 2004-2006 16.2 Energy-related CO2 emissions by region in the Reference Scenario 16.3 Per-capita energy-related CO2 emissions by region in the Reference Scenario 16.4 Transport sector energy-related CO2 emissions by region in the Reference Scenario 16.5 Emissions of major air pollutants by region in the Reference Scenario
List of boxes Part A: Global energy trends to 2030 Chapter 1. Context and principal assumptions 1.1 Improvements to the modelling framework in WEO-2008 1.2 Implications of new estimates of purchasing power parity
61 67
Chapter 2. Global energy trends 2.1 How do the new WEO assumptions affect global energy trends compared with last year?
79
Chapter 3. Oil market outlook 3.1 The impact on demand of removing oil subsidies 3.2 Averting a post-2010 oil-supply crunch
96 104
Chapter 4. Natural gas market outlook 4.1 Adequacy of gas-supply infrastructure and risks of shortages
116
Chapter 5. Coal market outlook 5.1 Alternative energy supplies: underground coal gasification and coal-mine methane
130
Chapter 7. Renewable energy outlook 7.1 Land use required to grow biomass for energy 7.2 Energy storage
161 169
Chapter 8. Energy use in cities 8.1 Methodological issues and city energy data
181
Chapter 9. Turning oil resources into reserves 9.1 Defining and measuring conventional and non-conventional hydrocarbon reserves in the WEO 9.2 Proposed new SEC reserve-reporting rules 9.3 Potential for exploiting Arctic resources 9.4 Pre-salt deepwater potential in Brazil 9.5 Deploying EOR: when, how and at what cost?
200 201 206 207 214
Chapter 10. Field-by-field analysis of oil production 10.1 The IEA field-by-field oil production database 10.2 Oilfield production-profiling methodology 10.3 How do we define and calculate decline rates? 10.4 The Ghawar field: the super-giant among super-giants
224 231 235 237
32
World Energy Outlook 2008
© OECD/IEA, 2008
Part B: Oil and gas production prospects
Chapter 11. Prospects for oil production 11.1 Modelling oil production in WEO-2008
252
Chapter 12. Natural gas resources and production prospects 12.1 Sour gas reserves: a costly nuisance? 12.2 CO2-enhanced coalbed methane (ECBM) production: a dual benefit?
283 286
Chapter 13. Upstream investment prospects 13.1 Approach to assessing the outlook for upstream investment 13.2 Impact of the credit crunch on upstream financing 13.3 IEA Upstream Investment Cost Index
305 309 318
Chapter 14. The structure of the upstream industry 14.1 The rise of the oilfield-services companies
341
Chapter 15. Prospects in oil- and gas-exporting sub-Saharan African countries 15.1 Gas flaring: what are the costs? 361 15.2 Regional power pools and the Grand Inga project 367 15.3 Government programmes to promote the use of clean cooking fuels and rural electrification in Africa: success stories 369 15.4 The prospects for renewables to meet rural household energy needs 372 15.5 Subsidies: how best to safeguard energy-poor households? 373 15.6 Revenue management in Botswana: a success story 375
Chapter 16. Implications of the Reference Scenario for the global climate 16.1 Projected energy-related CO2 emissions compared with WEO-2007 16.2 Global city energy-related CO2 emissions in the Reference Scenario 16.3 Greenhouse-gas emissions data and projections
383 390 397
Chapter 17. The post-2012 climate policy framework 17.1 Discerning a stabilisation limit from the IPCC findings 17.2 Principles for differentiating responsibilities 17.3 Non-binding targets for non-OECD countries? 17.4 Carbon leakage and loss of competitiveness 17.5 Sectoral approaches and the power-generation sector 17.6 CO2 savings potential of cities
411 419 427 429 431 432
Chapter 18. Climate policy scenarios 18.1 Outlook for energy-related CO2 emissions to 2050 18.2 Vintages of capital stock in iron and steel and cement sectors
447 461
Chapter 19. Implications for energy investment 19.1 Stepping up research, development and demonstration to promote climate-friendly technologies
491
Table of contents
33
© OECD/IEA, 2008
Part C: The role of energy in climate policy
19.2 19.3 19.4
Shutting down coal-fired power plants in Ontario, Canada What next for the Clean Development Mechanism? The importance of intellectual property laws for technology diffusion
493 497 500
List of spotlights Part A: Global energy trends to 2030 Chapter 1. Context and principal assumptions Are speculators to blame for soaring oil prices?
71
Chapter 3. Oil market outlook What is the scope for switching from oil-fuelled to electric vehicles?
101
Chapter 4. Natural gas market outlook Are high prices choking off demand for gas?
111
Chapter 6. Power sector outlook What are the risks of investing in the power sector?
155
Chapter 7. Renewable energy outlook To what extent are biofuels driving up the price of food?
173
Part B: Oil and gas production prospects Chapter 9. Turning oil resources into reserves How soon — if ever — could oil recovery factors be raised to 50%?
212
Chapter 10. Field-by-field analysis of oil production What do rising decline rates mean for oil production and investment?
223
Chapter 13. Upstream investment prospects Is strong demand leading to more drilling capacity?
320
Chapter 16. Implications of the Reference Scenario for the global climate How much of the world’s energy-related CO2 emissions are “embedded” in internationally traded goods and services?
386
Chapter 17. The post-2012 climate policy framework Why is it an absolute “must” for non-OECD countries to play their part in a global climate-change regime?
418
34
World Energy Outlook 2008
© OECD/IEA, 2008
Part C: The role of energy in climate policy
World Energy Outlook Series World Energy Outlook 1993 World Energy Outlook 1994 World Energy Outlook 1995 World Energy Outlook 1996 World Energy Outlook 1998 World Energy Outlook: 1999 Insights Looking at Energy Subsidies: Getting the Prices Right World Energy Outlook 2000 World Energy Outlook 2001 Insights Assessing Today’s Supplies to Fuel Tomorrow’s Growth World Energy Outlook 2002 World Energy Investment Outlook: 2003 Insights World Energy Outlook 2004 World Energy Outlook 2005 Middle East and North Africa Insights World Energy Outlook 2006 World Energy Outlook 2007 China and India Insights World Energy Outlook 2008 World Energy Outlook 2009 (forthcoming)
35
© OECD/IEA, 2008
More information available at www.worldenergyoutlook.org
© OECD/IEA, 2008
EXECUTIVE SUMMARY The world’s energy system is at a crossroads. Current global trends in energy supply and consumption are patently unsustainable — environmentally, economically, socially. But that can — and must — be altered; there’s still time to change the road we’re on. It is not an exaggeration to claim that the future of human prosperity depends on how successfully we tackle the two central energy challenges facing us today: securing the supply of reliable and affordable energy; and effecting a rapid transformation to a low-carbon, efficient and environmentally benign system of energy supply. What is needed is nothing short of an energy revolution. This World Energy Outlook demonstrates how that might be achieved through decisive policy action and at what cost. It also describes the consequences of failure.
Preventing catastrophic and irreversible damage to the global climate ultimately requires a major decarbonisation of the world energy sources. On current trends, energy-related emissions of carbon-dioxide (CO2) and other greenhouse gases will rise inexorably, pushing up average global temperature by as much as 6°C in the long term. Strong, urgent action is needed to curb these trends. The 15th Conference of the Parties, to be held in Copenhagen in November 2009, provides a vital opportunity to negotiate a new global climate-change policy regime for beyond 2012 (the final year of coverage of the first commitment period of the Kyoto Protocol). The conference will need to put in place a framework for long-term co-operative action to bring the world onto a well-defined policy path towards a clear, quantified global goal for the stabilisation of greenhouse gases in the atmosphere. It will also need to ensure broad participation and put in place robust policy mechanisms to achieve the agreed objective. Executive summary
37
© OECD/IEA, 2008
Oil is the world’s vital source of energy and will remain so for many years to come, even under the most optimistic of assumptions about the pace of development and deployment of alternative technology. But the sources of oil to meet rising demand, the cost of producing it and the prices that consumers will need to pay for it are extremely uncertain, perhaps more than ever. The surge in prices in recent years culminating in the price spike of 2008, coupled with much greater short-term price volatility, have highlighted just how sensitive prices are to short-term market imbalances. They have also alerted people to the ultimately finite nature of oil (and natural gas) resources. In fact, the immediate risk to supply is not one of a lack of global resources, but rather a lack of investment where it is needed. Upstream investment has been rising rapidly in nominal terms, but much of the increase is due to surging costs and the need to combat rising decline rates — especially in higher-cost provinces outside of OPEC. Today, most capital goes to exploring for and developing high-cost reserves, partly because of limitations on international oil company access to the cheapest resources. Expanding production in the lowest-cost countries will be central to meeting the world’s needs at reasonable cost in the face of dwindling resources in most parts of the world and accelerating decline rates everywhere.
The energy sector will have to play the central role in curbing emissions — through major improvements in efficiency and rapid switching to renewables and other lowcarbon technologies, such as carbon capture and storage (CCS). Securing energy supplies and speeding up the transition to a low-carbon energy system both call for radical action by governments — at national and local levels, and through participation in co-ordinated international mechanisms. Households, businesses and motorists will have to change the way they use energy, while energy suppliers will need to invest in developing and commercialising low-carbon technologies. To make this happen, governments have to put in place appropriate financial incentives and regulatory frameworks that support both energy-security and climate-policy goals in an integrated way. Removing subsidies on energy consumption, which amounted to a staggering $310 billion in the 20 largest non-OECD countries in 2007, could make a major contribution to curbing demand and emissions growth. High international oil prices, by deterring consumption and encouraging more efficient demand-side technologies, push in the same direction, but only at the expense of economic growth and of living standards in consuming countries, both rich and poor. And some of the alternatives to conventional oil that high prices encourage are even more carbon-intensive. Many countries have made progress in crafting national responses, but much more needs to be done. A new international climate agreement is but a first essential step on the road towards a sustainable energy system; effective implementation is just as crucial. Delay in doing either would increase the eventual cost of meeting any given global climate target.
More of the same: a vision of a laisser-faire fossil-energy future
Due to continuing strong economic growth, China and India account for just over half of the increase in world primary energy demand between 2006 and 2030. Middle East countries strengthen their position as an important demand centre, contributing a further 11% to incremental world demand. Collectively, non-OECD countries account for 87% of the increase. As a result, their share of world primary energy demand rises from 51% to 62%. Their energy consumption overtook that of the OECD in 2005. Global primary demand for oil (excluding biofuels) rises by 1% per year on average, from 85 million barrels per day in 2007 to 106 mb/d in 2030. However, its share 38
World Energy Outlook 2008
© OECD/IEA, 2008
In our Reference Scenario, world primary energy demand grows by 1.6% per year on average in 2006-2030, from 11 730 Mtoe to just over 17 010 Mtoe — an increase of 45%. To illustrate the course on which we are set, this scenario embodies the effects of those government policies and measures that were enacted or adopted up to mid-2008, but not new ones. This provides a baseline against which we can quantify the extent to which we need to change course. Demand grows at a slower rate than projected in WEO-2007, mainly due to higher energy prices and slower economic growth, especially in OECD countries. Fossil fuels account for 80% of the world’s primary energy mix in 2030 — down slightly on today. Oil remains the dominant fuel, though demand for coal rises more than demand for any other fuel in absolute terms. The share of the world’s energy consumed in cities — an estimated 7 900 Mtoe in 2006 — grows from two-thirds to almost three-quarters in 2030.
of world energy use drops, from 34% to 30%. Oil demand in 2030 has been revised downwards by 10 mb/d since last year’s Outlook, reflecting mainly the impact of much higher prices and slightly slower GDP growth, as well as new government policies introduced in the past year. All of the projected increase in world oil demand comes from non-OECD countries (over four-fifths from China, India and the Middle East); OECD oil demand falls slightly, due largely to declining non-transport oil demand. Global demand for natural gas grows more quickly, by 1.8% per year, its share in total energy demand rising marginally, to 22%. Most of the growth in gas use comes from the power-generation sector. World demand for coal advances by 2% a year on average, its share in global energy demand climbing from 26% in 2006 to 29% in 2030. Some 85% of the increase in global coal consumption comes from the power sector in China and India. The share of nuclear power in primary energy demand edges down over the Outlook period, from 6% today to 5% in 2030 (its share of electricity output drops from 15% to 10%), reflecting the consistency of our rule not to anticipate changes in national policies — notwithstanding a recent revival of interest in nuclear power. Nuclear output nonetheless increases in absolute terms in all major regions except OECD Europe. Modern renewable technologies grow most rapidly, overtaking gas to become the second-largest source of electricity, behind coal, soon after 2010. Falling costs as renewable technologies mature, assumed higher fossil-fuel prices and strong policy support provide an opportunity for the renewable industry to eliminate its reliance on subsidies and to bring emerging technologies into the mainstream. Excluding biomass, non-hydro renewable energy sources — wind, solar, geothermal, tide and wave energy — together grow faster than any other source worldwide, at an average rate of 7.2% per year over the projection period. Most of the increase occurs in the power sector. The share of non-hydro renewables in total power generation grows from 1% in 2006 to 4% in 2030. Hydropower output increases, though its share of electricity drops two percentage points to 14%. In the OECD, the increase in renewables-based power generation exceeds that in fossil-based and nuclear power generation combined.
The Reference Scenario projections call for cumulative investment of over $26 trillion (in year-2007 dollars) in 2007-2030, over $4 trillion more than posited in WEO-2007. The power sector accounts for $13.6 trillion, or 52% of the total. Most of the rest goes to oil and gas, mainly for exploration and development and mostly in non-OECD regions. Unit capital costs, especially in the oil and gas industry, have continued to surge in the last year, leading to an upward revision in our assumed costs for the projection period. That increase outweighs the slower projected expansion of the world energy system. The current financial crisis is not expected to affect longterm investment, but could lead to delays in bringing current projects to completion, particularly in the power sector. Just over half of projected global energy investment in 2007-2030 goes simply to maintain the current level of supply capacity: much of the world’s current infrastructure for supplying oil, gas, coal and electricity will need to be replaced by 2030. To provide adequate assurances about the circumstances that Executive summary
39
© OECD/IEA, 2008
Massive investments in energy infrastructure will be needed
will govern future investment in energy-supply infrastructure, negotiations need to be concluded urgently on an international agreement on combating climate change and the implications for national policies quickly assessed. These projections are based on the assumption that the IEA crude oil import price averages $100 per barrel (in real year-2007 dollars) over the period 2008-2015, rising to over $120 in 2030. This represents a major upward adjustment from last year’s Outlook, reflecting the higher prices for near-term physical delivery and for futures contracts, as well as a reassessment of the prospects for the cost of oil supply and the outlook for demand. In nominal terms, prices double to just over $200 per barrel in 2030. However, pronounced short-term swings in prices are likely to remain the norm and temporary price spikes or sharp falls cannot be ruled out. Prices are likely to remain highly volatile, especially in the next year or two. A worsening of the current financial crisis would most likely depress economic activity and, therefore, oil demand, exerting downward pressure on prices. Beyond 2015, we assume that rising marginal costs of supply exert upward pressure on prices through to the end of the projection period. Combined with our oil-demand projections, these assumptions point to persistently high levels of consumer spending on oil in both OECD and non-OECD countries. As a share of world GDP at market exchange rates, spending soared from 1% in 1998 to around 4% in 2007, with serious adverse implications for the economies of consuming countries. That share is projected to stabilise at more than 5% over much of the Outlook period. For non-OECD countries, the share averages 6% to 7%. The only time the world has ever spent so much of its income on oil was in the early 1980s, when it exceeded 6%. On the other hand, OPEC oil and gas export revenues jump from under $700 billion in 2006 to over $2 trillion in 2030, with their share of world GDP rising from 1.2% to 2%.
Most incremental oil and gas will come from OPEC – if they invest enough
The bulk of the increase in world oil output is expected to come from OPEC countries, their collective share rising from 44% in 2007 to 51% in 2030. Their reserves are, in principle, large enough (and development costs low enough) for output to grow faster than this. But investment by these countries is assumed to be constrained by several factors, including conservative depletion policies and geopolitics. Saudi Arabia remains the world’s largest producer throughout the projection period, its output climbing 40
World Energy Outlook 2008
© OECD/IEA, 2008
World oil supply is projected to rise from 84 mb/d in 2007 to 106 mb/d in 2030 in the Reference Scenario. Netting out processing gains in refining, global production reaches 104 mb/d. Although global oil production in total is not expected to peak before 2030, production of conventional oil — crude oil, natural gas liquids (NGLs) and enhanced oil recovery (EOR) — is projected to level off towards the end of the projection period. Conventional crude oil production alone increases only modestly over 2007-2030 — by 5 mb/d — as almost all the additional capacity from new oilfields is offset by declines in output at existing fields. The bulk of the net increase in total oil production comes from NGLs (driven by the relatively rapid expansion in gas supply) and from non-conventional resources and technologies, including Canadian oil sands.
from 10.2 mb/d in 2007 to 15.6 mb/d in 2030. Non-OPEC conventional oil production is already at plateau and is projected to start to decline by around the middle of the next decade, accelerating through to the end of the projection period. Production has already peaked in most non-OPEC countries and will peak in most others before 2030. Falling crude oil and NGLs production is largely offset by rising non-conventional output, which keeps total non-OPEC output broadly flat over the second half of the projection period. Conventional capacity, net of natural production declines at existing fields, is set to grow in the near term, but dwindling new discoveries and a fall in the size of new fields are expected to drive up marginal development costs, leading to a drop in output. The projected increase in global oil output hinges on adequate and timely investment. Some 64 mb/d of additional gross capacity — the equivalent of almost six times that of Saudi Arabia today — needs to be brought on stream between 2007 and 2030. Some 30 mb/d of new capacity is needed by 2015. There remains a real risk that under-investment will cause an oil-supply crunch in that timeframe. The current wave of upstream investment looks set to boost net oil-production capacity in the next two to three years, pushing up spare capacity modestly. However, capacity additions from current projects tail off after 2010. This largely reflects the upstream development cycle: many new projects will undoubtedly be sanctioned in the near term as oil companies complete existing projects and move on to new ones. But the gap now evident between what is currently being built and what will be needed to keep pace with demand is set to widen sharply after 2010. Around 7 mb/d of additional capacity (over and above that from all current projects) needs to be brought on stream by 2015, most of which will need to be sanctioned within the next two years, to avoid a fall in spare capacity towards the middle of the next decade. Production of natural gas is also set to become more concentrated in the most resource-rich regions. Some 46% of the projected growth in world gas production in 2006-2030 comes from the Middle East, its output tripling to around 1 trillion cubic metres (tcm) by 2030. About 60% of the region’s incremental output is consumed locally, mainly in power stations. Most of the remaining increase in world output is provided by Africa and Russia. If investments in these countries falter, lower gas supply could lead to greater reliance on coal and higher CO2 emissions.
The world’s total endowment of oil is large enough to support the projected rise in production beyond 2030 in the Reference Scenario. Estimates of remaining proven reserves of oil and NGLs range from about 1.2 to 1.3 trillion barrels (including about 0.2 trillion barrels of non-conventional oil). They have almost doubled since 1980. This is enough to supply the world with oil for over 40 years at current rates of consumption. Though most of the increase in reserves has come from revisions made in the 1980s in OPEC countries rather than from new discoveries, modest increases have continued since 1990, despite rising consumption. The volume of oil discovered each year on average has been higher since 2000 than in the 1990s, thanks to increased exploration activity and improvements in technology, though production continues to outstrip discoveries (despite some big recent finds, such as in deepwater offshore Brazil). Executive summary
41
© OECD/IEA, 2008
The world is not running short of oil or gas just yet
Ultimately recoverable conventional oil resources, which include initial proven and probable reserves from discovered fields, reserves growth and oil that has yet to be found, are estimated at 3.5 trillion barrels. Only a third of this total, or 1.1 trillion barrels, has been produced up to now. Undiscovered resources account for about a third of the remaining recoverable oil, the largest volumes of which are thought to lie in the Middle East, Russia and the Caspian region. Non-conventional oil resources, which have been barely developed to date, are also very large. Between 1 and 2 trillion barrels of oil sands and extra-heavy oil may be ultimately recoverable economically. These resources are largely concentrated in Canada (mainly in Alberta province) and Venezuela (in the Orinoco Belt). The total long-term potentially recoverable oil-resource base, including extra-heavy oil, oil sands and oil shales (another largely undeveloped, though costly resource), is estimated at around 6.5 trillion barrels. Adding coal-to-liquids and gas-toliquids increases this potential to about 9 trillion barrels. Globally, natural gas resources are large, but, like oil, are highly concentrated in a small number of countries and fields. Remaining proven reserves amount to 180 tcm — equal to around 60 years of current production. Three countries — Russia, Iran and Qatar — hold 56% of the world’s reserves, while just 25 fields worldwide hold almost half. OPEC countries also hold about half. Remaining reserves have more than doubled since 1980, with the biggest increases coming from the Middle East. Although the size of gas discoveries has been steadily declining in recent decades in the same way as for oil, discoveries continue to exceed production. Ultimately recoverable remaining resources of conventional natural gas, including remaining proven reserves, reserves growth and undiscovered resources, could amount to well over 400 tcm. Cumulative production to 2007 amounts to less than one-sixth of total initial resources. Nonconventional gas resources — including coalbed methane, tight gas sands and gas shales — are much larger, amounting perhaps to over 900 tcm, with 25% in the United States and Canada combined.
Globally, oil resources might be plentiful, but there can be no guarantee that they will be exploited quickly enough to meet the level of demand projected in our Reference Scenario. One major uncertainty concerns the rate at which output from producing oilfields declines as they mature. This is a critical determinant of the amount of new capacity and investment that will be needed globally to meet projected demand. The findings of a detailed field-by-field analysis of the historical production trends of 800 fields, set out in Part B of this Outlook, indicate that observed decline rates (the observable fall in production) are likely to accelerate in the long term in each major world region. This results from a fall in the average size of field and, in some regions, an increase in the share of production that is expected to come from offshore fields. Our analysis demonstrates that, in general, the larger a field’s reserves, the lower the peak relative to reserves and the slower the decline once the field has passed its peak. Rates are also lower for onshore than offshore (especially deepwater) fields. Investment and production policies also affect decline rates. 42
World Energy Outlook 2008
© OECD/IEA, 2008
But field-by-field declines in oil production are accelerating…
We estimate that the average production-weighted observed decline rate worldwide is currently 6.7% for fields that have passed their production peak. In our Reference Scenario, this rate increases to 8.6% in 2030. The current figure is derived from our analysis of production at 800 fields, including all 54 super-giants (holding more than 5 billion barrels) in production today. For this sample, the observed post-peak decline rate averaged across all fields, weighted by their production over their whole lives, was found to be 5.1%. Decline rates are lowest for the biggest fields: they average 3.4% for super-giant fields, 6.5% for giant fields and 10.4% for large fields. Observed decline rates vary markedly by region; they are lowest in the Middle East and highest in the North Sea. This reflects, to a large extent, differences in the average size of fields, which in turn is related to the extent to which overall reserves are depleted and whether they are located onshore or offshore. Adjusting for the higher decline rates of smaller fields explains the higher estimated decline rate for the world, compared with that based on our dataset. Natural, or underlying, decline rates are about a third higher on average than observed decline rates, though the difference varies across regions reflecting differences in investment. (The natural decline rate strips out the effects of ongoing and periodic investment.) For the world as a whole, it is estimated at 9% for post-peak fields. In other words, the decline in production from existing fields would have been around one-third faster had there been no capital spending on those fields once they had passed their peak. Our Reference Scenario projections imply an increase in the global average natural decline rate to around 10.5% per year by 2030 (almost two percentage points higher than the observed rate), as all regions experience a drop in average field size and most see a shift in production to offshore fields over the projection period. This means that total upstream investment in some countries will need to rise, in some cases significantly, just to offset this faster decline. The implications are far-reaching: investment in 1 mb/d of additional capacity — equal to the entire capacity of Algeria today — is needed each year by the end of the projection period just to offset the projected acceleration in the natural decline rate.
Faster natural decline rates will mean a need for more upstream investment, both in existing fields (to combat natural decline) and in new fields (to offset falling production from existing fields and to meet rising demand). In fact, total upstream investment (in oil and gas fields) has been rising rapidly in recent years, more than tripling between 2000 and 2007 to $390 billion in nominal terms. Most of this increase was to meet higher unit costs: in cost-inflation adjusted terms, investment in 2007 was 70% higher than in 2000. Worldwide, upstream costs rose on average by an estimated 90% between 2000 and 2007 and by a further 5% in the first half of 2008, according to the IEA Index of Upstream Capital Costs. Most of the increase occurred in 2004-2007. Based on the plans of 50 of the world’s largest companies surveyed for this Outlook (accounting for more than three-quarters of world oil and gas production), global upstream oil and gas investment is expected to continue to rise, to just over $600 billion in nominal terms by 2012 — an increase of more than half over 2007. If costs level off, as assumed, real spending in the five years to 2012 would grow by 9% per year — about the same rate as in the previous seven years. Executive summary
43
© OECD/IEA, 2008
…and barriers to upstream investment could constrain global oil supply
The Reference Scenario projections imply a need for cumulative investment in the upstream oil and gas sector of around $8.4 trillion (in year-2007 dollars) over 2007-2030, or $350 billion per year on average. That is significantly less than is currently being spent. This is due to a major shift in where that investment is needed. Much more capital needs to go to the resource-rich regions, notably the Middle East, where unit costs are lowest. In short, the opportunities for international companies to invest in non-OPEC regions will diminish as the resource base contracts, eventually leaving the countries holding the bulk of the world’s remaining oil and gas reserves to take on a larger burden of investment, either directly through their national companies or indirectly, in partnership with foreign investors. It cannot be taken for granted that these countries will be willing to make this investment themselves or to attract sufficient foreign capital to keep up the necessary pace of investment.
Stronger oil company partnerships could bring mutual benefits
How the structure of the global oil and gas industry evolves in the coming decades will have important implications for investment, production capacity and prices. The increasing dominance of national companies may make it less certain that the investment projected in this Outlook will actually be made. The long-term policies of some major resource-rich countries in support of national goals may lead to slower depletion of their resources. Although some national companies, like Saudi Aramco, perform strongly in most areas, others may be less well placed to address the financing, technical and managerial challenges of bringing new upstream capacity on stream. Partnerships between the national and international companies could help address these challenges. The mutual benefits that could accrue are compelling: the national companies control most of the world’s remaining reserves, but some lack the technology and skilled personnel to do much more than simply maintain existing producing assets; the international companies are opportunity-constrained, but have the management skills and technology to help national companies develop their reserves. 44
World Energy Outlook 2008
© OECD/IEA, 2008
Major structural changes are underway in the upstream oil and gas industry, with the national companies playing an increasingly dominant role. In the Reference Scenario, they account for about 80% of total incremental production of both oil and gas between 2007 and 2030. In most of the countries with the largest oil and gas reserves, national companies dominate the upstream industry and foreign companies are either not allowed to own and develop reserves or are subject to tight restrictions. Higher oil prices and a growing conviction among political leaders that national companies serve the nation’s interests better than private and foreign oil companies have boosted the confidence and aspirations of national companies, some of them rivalling the international companies for technical capability and efficiency. The international oil companies, which have traditionally dominated the global oil and gas industry, are increasingly being squeezed by the growing power of the national companies and by dwindling reserves and production in accessible mature basins outside OPEC countries. The super-majors have been struggling to replace their proven reserves and expand production, while the share of their cash earnings that is returned to shareholders has been growing.
Oil-rich African countries have no excuse for their citizens’ energy poverty A number of sub-Saharan African countries hold large oil and gas resources, which are expected to underpin strong growth in their production and exports in the coming two decades or so. Conventional oil production in the ten largest hydrocarbon-producing countries in sub-Saharan Africa reached 5.6 mb/d in 2007, of which 5.1 mb/d was exported. In the Reference Scenario, output rises to 7.4 mb/d and oil exports to 6.4 mb/d in 2030. Gas production in these countries increases more than four-fold, from 36 billion cubic metres in 2006 to 163 bcm in 2030, with most of the increase going to export. These projections hinge on a reduction in gas flaring, adequate investment and avoiding disruptions to supplies through civil unrest. Cumulative government revenues from oil and gas output (from royalties and taxes) in these ten countries are projected, in aggregate, to total $4 trillion over 2007-2030. Nigeria and Angola remain the largest exporters, with combined cumulative government revenues of about $3.5 trillion. Taxes on oil and gas production account for more than 50% of total government revenues in most of the oil- and gas-rich sub-Saharan African countries. Despite the vast hydrocarbon wealth of these ten countries, most of their citizens remain poor. As a result, household use of modern energy services is very limited. Twothirds of households do not have access to electricity and three-quarters do not have access to clean fuels for cooking, relying instead on fuelwood and charcoal. Unless there are major government initiatives to address this problem, the number of electricitydeprived people is projected to increase over the projection period, as the population grows. And more than half of the total population of these countries still relies on fuelwood and charcoal for cooking in 2030. Tackling energy poverty is well within these countries’ means, but major institutional reforms are needed. We estimate the capital cost of providing minimal energy services (electricity and liquefied petroleum gas stoves and cylinders) to these households over the Outlook period to be about $18 billion. This is equivalent to only 0.4% of cumulative government revenues from oil and gas. An improvement in the efficiency and transparency of revenue allocation and the accountability of governments in the use of public funds would improve the likelihood that oil and gas revenues are actually used to alleviate poverty generally and energy poverty specifically.
The projected rise in emissions of greenhouse gases in the Reference Scenario puts us on a course of doubling the concentration of those gases in the atmosphere by the end of this century, entailing an eventual global average temperature increase of up to 6°C. The Reference Scenario trends point to continuing growth in emissions of CO2 and other greenhouse gases. Global energy-related CO2 emissions rise from 28 Gt in 2006 to 41 Gt in 2030 — an increase of 45%. The 2030 projection is only 1 Gt lower than that projected in last year’s Outlook, even though we assume much higher prices and slightly lower world GDP growth. World greenhouse-gas emissions, including nonExecutive summary
45
© OECD/IEA, 2008
The consequences for the global climate of policy inaction are shocking
energy CO2 and all other gases, are projected to grow from 44 Gt CO2-equivalent in 2005 to 60 Gt CO2-eq in 2030, an increase of 35% over 2005. Three-quarters of the projected increase in energy-related CO2 emissions in the Reference Scenario arises in China, India and the Middle East, and 97% in non-OECD countries as a whole. On average, however, non-OECD per-capita emissions remain far lower than those in the OECD. Emissions in the OECD reach a peak after 2020 and then decline. Only in Europe and Japan are emissions in 2030 lower than today. The bulk of the increase in global energy-related CO2 emissions is expected to come from cities, their share rising from 71% in 2006 to 76% in 2030 as a result of urbanisation. City residents tend to consume more energy than rural residents, so they therefore emit more CO2 per capita.
The road from Copenhagen must be paved with more than good intentions Strong, co-ordinated action is needed urgently to curb the growth in greenhousegas emissions and the resulting rise in global temperatures. The post-2012 global climate-change policy regime that is expected to be established at the Copenhagen conference in 2009 will provide the international framework for that action. With energy-related CO2 accounting for 61% of global greenhouse-gas emissions, the energy sector will have to be at the heart of discussions on what level of concentration to aim for and how to achieve it. The target that is set for the long-term stabilisation of greenhouse-gas concentration will determine the pace of the required transformation of the global energy system, as well as how stringent the policy responses will need to be. Successfully meeting that target will hinge on effective implementation.
Any agreement will need to take into account the importance of a handful of major emitters. The five largest emitters of energy-related CO2 — China, the United States, the European Union, India and Russia — together account for almost two thirds of 46
World Energy Outlook 2008
© OECD/IEA, 2008
The choice of global emissions trajectory will need to take into account technological requirements and costs in the energy sector. The normal cycle of capital replacement is a key constraint on the speed with which low-carbon technologies can enter into use without incurring disproportionate cost. The energy sector has a relatively slow rate of capital replacement in general, due to the long lifetime of much of its capital — for producing, supply and using energy. As a result, more efficient technologies normally take many years to spread through the energy sector. It will be necessary to face up to the reality of the cost of early capital retirement if radical measures are to be taken to speed up this process so as to deliver deep cuts in emissions. The rate of capital-stock turnover is particularly slow in the power sector, where large up-front costs and long operating lifetimes mean that plants that have already been built — and their associated emissions — are effectively “locked-in”. In the Reference Scenario, three-quarters of the projected output of electricity worldwide in 2020 (and more than half in 2030) comes from power stations that are already operating today. As a result, even if all power plants built from now onwards were carbon-free, CO2 emissions from the power sector would still be only 25%, or 4 Gt, lower in 2020 relative to the Reference Scenario.
global CO2 emissions; in the Reference Scenario, this proportion is expected to remain similar in 2020. The contributions to emissions reduction made by China and the United States will be critical to reaching a stabilisation goal. The scale of the reduction in energy-related emissions by country or region varies markedly with different levels of international participation.
The stabilisation goal will determine the scale of the energy challenge This Outlook considers two climate-policy scenarios corresponding to long-term stabilisation of greenhouse-gas concentration at 550 and 450 parts per million of CO2 equivalent. The 550 Policy Scenario equates to an increase in global temperature of approximately 3°C, the 450 Policy Scenario to a rise of around 2°C. The 550 Policy Scenario involves a plateauing of greenhouse-gas emissions by 2020 and reductions soon after. The 450 Policy Scenarios involves much more substantial reductions after 2020. Even then, emissions overshoot the trajectory needed to meet the 450 ppm CO2-eq target, requiring greater emissions reductions after 2030. In both scenarios, total emissions are significantly lower in 2030 in all major emitting countries. To reach either of these outcomes, hundreds of millions of households and businesses around the world would need to be encouraged to change the way they use energy. This will require innovative policies, an appropriate regulatory framework, the rapid development of a global carbon market and increased investment in energy research, development and demonstration.
In the 550 Policy Scenario, world primary energy demand expands by about 32% between 2006 and 2030 with the share of fossil fuels falling markedly. Demand grows on average by 1.2% per year, compared with 1.6% in the Reference Scenario. By 2030, demand is 9% lower than in the Reference Scenario, mainly as a result of efficiency gains. Global energy-related CO2 emissions peak in 2025 and then decline slightly to 33 Gt in 2030, while greenhouse-gas emissions plateau by 2020 and are broadly flat through to 2030. Both total greenhouse-gas and energy-related CO2 emissions are 19% lower in 2030 than in the Reference Scenario. The energy mix in this scenario is markedly different to that of the Reference Scenario, with fossil fuels losing market share to renewables and nuclear power. Oil demand rises to 98 mb/d in 2030 — almost 9 mb/d less than in the Reference Scenario. More than half of the oil savings occur in the transport sector in OECD countries and other major economies, as a result of Executive summary
47
© OECD/IEA, 2008
There is a wide range of international policy mechanisms that could be adopted to meet an agreed climate objective. However, as current political debate shows, and given practical issues in the energy sector, the reality is that nations adopt the approach or approaches that best reflect their varied interests and capabilities. This Outlook analyses the implications for the energy sector of a hybrid policy framework involving one particular combination of cap-and-trade systems, sectoral agreements (in the transport and industry sectors) and national policies and measures. Cap-andtrade systems are assumed to play an important role in the OECD regions. The carbon price there reaches $90/tonne of CO2 in 2030 in the 550 Policy Scenario and $180/tonne in the 450 Policy Scenario.
sectoral agreements to reduce emissions from light-duty vehicles and aviation. Oil prices are around $100 per barrel (in year-2007 dollars) in 2030, 18% lower than in the Reference Scenario, due to lower demand. OPEC production still increases, to 49 mb/d in 2030, almost 13 mb/d higher than today (but 4 mb/d less than in the Reference Scenario). CCS is also deployed more quickly. In 2030, the installed capacity of CCS plants amounts to over 160 GW worldwide, of which about 70% is in OECD countries. CCS capacity is negligible in the Reference Scenario. The 450 Policy Scenario assumes much stronger and broader policy action from 2020 onwards, inducing quicker development and deployment of low-carbon technologies. Global energy-related CO2 emissions are assumed to follow broadly the same trajectory as in the 550 Policy Scenario until 2020, and then to fall more quickly. They peak in 2020 at 32.5 Gt and then decline to 25.7 Gt in 2030. This scenario requires emissions in OECD countries to be reduced by almost 40% in 2030, compared with 2006 levels. Other major economies are required to limit their emissions growth to 20%. Participation in an international cap-and-trade system is assumed to be broader than in the 550 Policy Scenario, covering all major emitting countries from 2020 onwards. Hydropower, biomass, wind and other renewables see faster deployment in power generation, accounting for 40% of total generation worldwide in 2030. An additional 190 GW of CCS is deployed in the last decade of the projection period compared with the 550 Policy Scenario. The scale of the challenge in the 450 Policy Scenario is immense: the 2030 emissions level for the world as a whole in this scenario is less than the level of projected emissions for non-OECD countries alone in the Reference Scenario. In other words, the OECD countries alone cannot put the world onto the path to 450-ppm trajectory, even if they were to reduce their emissions to zero. Even leaving aside any debate about the political feasibility of the 450 Policy Scenario, it is uncertain whether the scale of the transformation envisaged is even technically achievable, as the scenario assumes broad deployment of technologies that have not yet been proven. The technology shift, if achievable, would certainly be unprecedented in scale and speed of deployment. Increased public and private spending on research and development in the near term would be essential to develop the advanced technologies needed to make the 450 Policy Scenario a reality.
The profound shifts in energy demand and supply in the two climate-policy scenarios call for huge increases in spending on new capital stock, especially in power plants and in more energy-efficient equipment and appliances. The 550 Policy Scenario requires $4.1 trillion more investment in total between 2010 and 2030 than in the Reference Scenario — equal on average to 0.24% of annual world GDP. Most of this goes to deploying and improving existing technologies. Investment in power plants is $1.2 trillion higher, with close to three-quarters of this additional capital going to OECD countries. Additional expenditures on the demand side are even bigger. Most of the extra spending is by individuals, who have to pay more for more efficient cars, appliances and buildings. This extra cost amounts to $17 per person per year 48
World Energy Outlook 2008
© OECD/IEA, 2008
Tackling climate change will require big shifts in spending
on average throughout the world. But these investments are accompanied by large savings on energy bills. Improved energy efficiency lowers fossil-fuel consumption by a cumulative amount of 22 billion tonnes of oil equivalent over 2010-2030, yielding cumulative savings of over $7 trillion. The additional up front expenditures on energy-related capital are, unsurprisingly, considerably larger in the 450 Policy Scenario. An additional $2.4 trillion needs to be invested in low- or zero-carbon power-generation capacity and an additional $2.7 trillion invested in more energy-efficient equipment, appliances and buildings than in the 550 Policy Scenario. Together, these costs equal on average 0.55% of annual world GDP. These expenditures are particularly great during the last decade of the projection period, when CO2 emissions fall most rapidly and the marginal cost of abatement options rises sharply. Galvanising these investments would require clear price signals (including through a broad-based, efficient carbon market), appropriate fiscal incentives and well-targeted regulation. At $5.8 trillion, the cumulative savings on fuel bills are smaller than in the 550 Policy Scenario, because higher electricity prices offset the bigger energy savings.
The energy future will be very different
Executive summary
49
© OECD/IEA, 2008
For all the uncertainties highlighted in this report, we can be certain that the energy world will look a lot different in 2030 than it does today. The world energy system will be transformed, but not necessarily in the way we would like to see. We can be confident of some of the trends highlighted in this report: the growing weight of China, India, the Middle East and other non-OECD regions in energy markets and in CO2 emissions; the rapidly increasing dominance of national oil companies; and the emergence of low-carbon energy technologies. And while market imbalances could temporarily cause prices to fall back, it is becoming increasingly apparent that the era of cheap oil is over. But many of the key policy drivers (not to mention other, external factors) remain in doubt. It is within the power of all governments, of producing and consuming countries alike, acting alone or together, to steer the world towards a cleaner, cleverer and more competitive energy system. Time is running out and the time to act is now.
© OECD/IEA, 2008
INTRODUCTION
Global energy at a crossroads Purpose and scope of the study It is not an exaggeration to say that the future of human prosperity hinges on finding a way of supplying the world’s growing energy needs in a way that does not irreparably harm the environment. Until recently, it looked as if we had plenty of time to meet that challenge. No longer. Surging oil and gas prices have drawn attention to the physical and political constraints on raising production — and the vital importance of affordable supplies to the world economy. And the latest scientific evidence suggests that the pace of climate change resulting from man-made emissions of greenhouse gases — the bulk of which come from burning fossil fuels — is faster than predicted. The urgent need for a veritable energy revolution, involving a wholesale global shift to low-carbon technologies, is now widely recognised. This year’s World Energy Outlook, as in all even-numbered years, provides a comprehensive update of long-term energy demand and supply projections to 2030, fuel by fuel and region by region. As a special feature this year, it examines how energy is used in cities and the implications for the future. The Outlook also takes a detailed look at the two headline issues that emerge from these projections: the prospects for oil and gas production, and the policy options for tackling climate change after 2012, when a new global agreement is due to take effect. The results of these analyses are intended to provide policy makers, investors and end users with a rigorous quantitative framework for assessing likely future trends in energy markets and the costeffectiveness of new policies to tackle energy-security and environmental concerns.
To shed light on these issues, this Outlook takes an in-depth look at the drivers of oil production, based on a detailed field-by-field analysis of production trends, investment and costs, and considers the opportunities for expanding capacity and the possible constraints and risks — both above and below ground. The aim is to support the cause of Introduction
51
© OECD/IEA, 2008
Last year’s Outlook drew attention to the risk of an oil-supply crunch by the middle of the next decade, were upstream investment to fall short of requirements, triggering a jump in prices. The surge in oil prices over the past year reflects, in large part, market expectations that such a scenario is set to become reality. A growing number of oil companies and analysts have suggested that oil production may peak within the next two decades, as a result of limits on the amount of investment that can be mobilised, rising costs and political and geological factors. Limited exploration activity is increasing the uncertainties about finding new oilfields; and the rate of decline in production at existing fields — especially the large, mature fields that have been the mainstay of global output for several decades — is faster than many believe. How decline rates at these fields evolve — with or without the deployment of enhanced recovery techniques — will have major implications for the need to invest in new fields, which are typically smaller, more complex and more costly to develop.
more transparency in data on oil and gas reserves and production; and provide valuable insights into the adequacy of current investment plans and the prospects for expanding output in the longer term. Worries about oil-supply shortages are matched by long-term concerns about the threat of rising fossil-energy use on the global climate. In most countries, climatechange has risen to the top of the political agenda — the result of a growing body of evidence of actual global warming and ever-more startling predictions of the ecological consequences. Action in the energy sector, the major source of man-made greenhousegas emissions, will be fundamental to meeting the climate-change challenge. Much depends on government responses. Past editions of the Outlook showed that even if all the government policies and measures under discussion at that time had been implemented, the world would still not be on a sustainable path. Since the last Outlook, governments have adopted significant new measures. But limiting global warming to a maximum of 2°C will require a much bigger and co-ordinated global effort. Negotiations on an international agreement to succeed the Kyoto Protocol are scheduled to end in 2009 with a meeting of the Conference of the Parties of the UN Framework Convention on Climate Change in Copenhagen. This year’s report aims to inform and support the climate negotiations by presenting information and analysis of a combination of policy options and approaches available to policy makers and their implications for the energy sector.
Methodology As in previous Outlooks, a scenario approach is used to examine future energy trends. The projection period runs to 2030. The projections are derived from a large-scale mathematical model, the World Energy Model (WEM), which has been completely overhauled and updated, drawing on the most recent historical data and revised assumptions. The core projections, the Reference Scenario, indicate what would happen if, among other things, there were to be no new energy-policy interventions by governments beyond those already adopted by mid-2008. This will not happen and the Reference Scenario is not a forecast: it is a baseline picture of how global energy markets would evolve if the underlying trends in energy demand and supply are not changed. This allows us to test alternative assumptions about future government policies. Additional scenarios and cases are included in the analyses of oil-production prospects and climate-policy options, as described below.
A central pillar of our analysis of the prospects for oil production to 2030 is a detailed field-by-field assessment of the production outlook for the world’s largest oilfields currently in production, the biggest of which, in most cases, have been in operation for many years. The analysis covers around 800 oilfields. To our knowledge, this is the most detailed and comprehensive study of its kind that has ever been made public. The work also involved a bottom-up assessment of near-term investment plans and capacity additions at existing and new oil and gas fields, based on the compilation of detailed project-by-project and company-by-company data. In recognition of the 52
World Energy Outlook 2008
© OECD/IEA, 2008
Oil and gas supply prospects
uncertainties surrounding the outlook for sustaining oil production at existing fields and fields that will be developed in the future, we have modelled the sensitivity of production to different production-decline rates and drawn out the implications for upstream investment and international oil markets. This analysis also considers the prospects for further advances in upstream technologies, the potential for improved and enhanced recovery techniques to bolster conventional oil and gas output and recovery rates, and the prospects for producing oil and gas from non-conventional sources. It also looks at the factors that may hinder capital flows to the upstream industry and capacity additions — including limitations on the opportunities for the international oil companies to invest in developing reserves in many key resource-rich countries and the willingness and ability of the resource-rich countries and their national companies to expand capacity. Higher prices have driven up the revenues earned by oil and gas exporters, which may paradoxically reduce the incentive for some countries to expand capacity in the medium term — partly because of limits on the rate at which they can absorb the additional earnings. The implications for the relationships between international oil companies, national companies and oil services providers are also assessed. As a special focus, we also look at the outlook for oil and gas production in ten oil- and gas-exporting sub-Saharan African countries, the revenues it will generate and the prospects for energy poverty. It is well established that the reasons for the persistence of poverty in resource-exporting countries are mismanagement of hydrocarbon revenues, poor governance and a lack of transparency. This study shows what could be achieved to end reliance on traditional biomass and provide general access to electricity by calculating the investment costs associated with increasing energy access and comparing them with total government revenues from hydrocarbon exploitation.
Climate change scenarios
The mix of approaches differs between three groups of countries: “OECD+” countries (a group that includes the OECD and non-OECD EU countries), “Other Major Economies” (including China, Russia, India, Indonesia, Brazil and Middle East) and “Other Countries”. A cap-and-trade system, similar in concept to the Emissions Trading Scheme that is currently operating in the European Union, is assumed to be implemented by the OECD+ countries in the 550 Policy Scenario and in the Other Major Economies too in the 450 Policy Scenario. The system is assumed to cover power generation and the industrial sector. The Other Major Economies are assumed to introduce national Introduction
53
© OECD/IEA, 2008
We examine in detail, through scenario analysis, how different international agreements and national/regional commitments on climate change could affect the evolution of energy markets after 2012. The analysis focuses on three of the most commonly discussed elements of a post-2012 climate framework, namely cap-andtrade, international sectoral agreements and national policies and measures. A plausible combination of these mechanisms, or hybrid approach, is modelled, using the WEM, in two scenarios: one in which it is agreed to limit the greenhouse-gas concentration in the atmosphere to 550 ppm (in CO2-equivalent terms) and another in which the concentration is held to 450 ppm.
policies and measures covering power generation and industry, aimed generally at improving energy security and mitigating local pollution, as well as helping to reduce CO2 emissions. In addition, to limit the risk of carbon leakage from countries covered by the cap-and-trade system, sector-specific energy-efficiency standards are assumed to be introduced in a number of industrial sectors. Achieving abatement of emissions in these countries is essential to meet ambitious climate targets, so these countries must have strong incentives to act, even for the 550 ppm objective to be met.
54
World Energy Outlook 2008
© OECD/IEA, 2008
We have assumed that international sectoral agreements are adopted and implemented in the transport and industry sectors in OECD+ and Other Major Economies. Such an approach is already being taken by the European Union, which currently excludes road transport from its emissions-trading scheme and is instead negotiating vehicle fuelefficiency standards and renewable fuel mandates. In the buildings sector, it is assumed that all countries undertake national policies and measures tailored to the national potential for energy savings. These could require formal international approval. It is assumed that the least-developed countries with low per-capita emissions are not required to play any part in a cap-and-trade system or adopt sectoral agreements. Nonetheless, they do act to limit emissions through national policies and measures, and have opportunities to earn emission credits under the Clean Development Mechanism and gain access to global technology or financing programmes.
© OECD/IEA, 2008
© OECD/IEA, 2008
PART A GLOBAL ENERGY TRENDS TO 2030
PREFACE
Higher energy prices and slower economic growth have left their mark on the assumptions for this year’s World Energy Outlook. As a result, energy growth to 2030 is slower than in WEO-2007, though other trends are broadly unchanged – persistent dominance of fossil fuels in the energy mix (despite additional action taken in the last twelve months to mitigate climate change), an inexorable rise in CO2 emissions and a strongly rising share for developing countries in global energy consumption: almost all the increase in fossil fuel consumption over the period occurs in non-OECD countries. Following the pattern which will be familiar to regular WEO readers, this first part of the book, Part A, sets out in detail this year’s world energy projections in the Reference Scenario. The main assumptions underpinning these projections are set out in Chapter 1. Chapter 2 summarises the global trends in energy demand and supply, as well as the implications for investment. The detailed projections for oil, gas, coal, electricity and renewables are then set out in Chapters 3-7.
© OECD/IEA, 2008
In an innovation this year, Chapter 8 examines the special place of urban energy consumption in global usage. By 2030, urban use accounts for nearly three-quarters of total energy use worldwide.
© OECD/IEA, 2008
CHAPTER 1
CONTEXT AND PRINCIPAL ASSUMPTIONS H
I
G
H
L
I
G
H
T
S
The IEA crude oil import price, a proxy for international prices, is assumed to average $100 per barrel in real year-2007 dollars over the period 2008-2015 and then to rise in a broadly linear manner to over $120 in 2030. This represents a major upward adjustment from last year’s Outlook, reflecting higher prices for near-term physical delivery and for futures contracts, as well as a reassessment of the prospects for the cost of oil supply and the outlook for demand. In nominal terms, prices double to just over $200 per barrel in 2030. Natural gas prices are assumed to rise in 2008 and then to fall back slightly through to 2010 in lagged adjustment to oil prices. Gas prices begin to rise after 2015 in line with oil prices. Coal prices, which have also risen sharply in recent years, are assumed to settle at around $120 per tonne in 2010, remaining flat through to 2015 and then falling back slightly to $110 in 2030. The average energy efficiency of equipment currently in use is generally assumed to improve progressively over the projection period, though at varying rates according to the type of equipment and the rate of retirement and replacement of the stock of capital. Some major new supply-side technologies, including carbon capture and storage, second-generation biofuels and coal-to-liquids, are assumed to be deployed on a small scale before the end of the projection period.
Chapter 1 - Context and principal assumptions
59
© OECD/IEA, 2008
The Reference Scenario embodies the effects of those government policies and measures that were enacted or adopted by mid-2008, even though many of them have not yet been fully implemented. Possible, potential or even likely future policy actions are not taken into account. Many countries subsidise energy, in some cases heavily. Most have policies to phase out subsidies, and are accordingly assumed to reduce them progressively. World population is assumed to grow at an annual average rate of 1%, from an estimated 6.5 billion in 2006 to 8.2 billion in 2030. Population growth slows progressively over the projection period, in line with past trends: population expanded by 1.4% per year from 1990 to 2006. The population of non-OECD countries as a group continues to grow most rapidly. The rate of growth in world GDP — the main driver of energy demand in all regions — is assumed to average 3.3% per year over the period 2006-2030. It averaged 3.2% from 1990 to 2006. The increase reflects the rising weight in the world economy of the non-OECD countries, where growth will remain fastest. The average growth rate nonetheless falls progressively over the projection period, from 4.2% in 2006-2015 to 2.8% in 2015-2030. Growth remains highest in China, India and the Middle East.
Government policies and measures As in previous editions of the World Energy Outlook, the Reference Scenario, the results of which are set out in Part A, takes into consideration all government policies and measures that were enacted or adopted by the middle of the current year, including those that have not yet been fully implemented. The policies considered cover a wide array of sectors and a variety of instruments. Some measures not directly addressed at the energy sector nonetheless have important consequences for energy markets and have been taken into account. Most recent policy initiatives directed at the energy sector, in both OECD and non-OECD countries, are designed to improve energy security, combat climate change and address other environmental problems through improved energy efficiency, switching to lower carbon fuels and enhanced indigenous energy production. They include the Energy Independence and Security Act in the United States and the European Commission’s climate action plan (the provisions of which are taken into account in the Reference Scenario), which includes a binding target to raise the share of renewable energy in the European Union’s primary consumption to 20% by 2030 (Table 1.1). The impact on energy demand and supply of recently adopted measures does not show up in historical market data, which are available only up to 2006 for all countries (preliminary data for 2007 is to hand for OECD countries). Some policies adopted earlier are not yet fully implemented but have been taken into account in the Reference Scenario. Table 1.1
Major new energy-related policy initiatives adopted between mid-2007 and mid-2008
Country/region
Policy/measure
Implementation in the Reference Scenario
European Union
European Commission integrated energy and climate action plan
$30/t CO2 price (in year-2007 prices) in the EU trading scheme sectors is assumed Increased use of renewable energy Improved energy efficiency Increased use of biofuels in transport
United States
Energy Independence and Security Act
Tightened corporate average fuel-economy standards, resulting in lower transport-fuel intensity Increase in biofuels consumption Tougher appliance standards and incentives result in lower energy intensity in buildings and industry
Japan
Revision of Act on the Rational Use of Energy
Improved energy efficiency and reduced CO2 emissions in the residential/services and other sectors (both measures)
Revision of Law on Global Warming Countermeasures Renewable Energy Development Plan Increased use of renewable energy Revision of Law on Energy Conservation
Improved energy efficiency
Note: More details about the policies in the Reference Scenario can be found at the WEO website www.worldenergyoutlook.org.
60
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
China
1
Importantly, the Reference Scenario does not include possible, potential or even likely future policy initiatives. For that reason, the projections in this scenario cannot be considered forecasts of what is likely to happen. Rather, they should be seen as a baseline vision of how energy markets are likely to develop should government policy making develop no further. Major new energy-policy initiatives will inevitably be implemented during the projection period, but it is impossible to predict exactly which measures will eventually be adopted in each country and how they will be implemented, especially towards the end of the projection period. The climate-policy scenarios set out in Part C analyse the impact of a range of possible future policy interventions. Box 1.1
Improvements to the modelling framework in WEO-2008
The IEA’s World Energy Model (WEM) — a large-scale mathematical construct designed to replicate how energy markets function — is the main tool used to generate the sector-by-sector and fuel-by-fuel projections by region or country for all the scenarios in this Outlook.1 The model, which has been developed over several years, comprises five main modules: final energy demand, power generation, refining and other transformation, investment and carbon-dioxide (CO2) emissions, all of which are modelled on a country or regional basis; and fossil-fuel and biofuels supply, which is modelled by major producer. This year, all the WEM modules were overhauled and major new elements were introduced, including the following: The demand module for all regions was completely rebuilt, involving re-estimating parameters using more recent time-series data and introducing more detailed coverage of demand by sector and fuel. The oil and gas production and trade models were expanded to take better account of economic variables and to reflect the recent surge in cost inflation and the fall in the value of the US dollar against most other currencies. Oilfield decline rates were analysed in detail on a field-by-field basis in order to assess the prospects for future decline rates (see Chapter 10). Biofuels demand was modelled in more detail, taking account of technical and economic factors. Power-generation capital and operating costs were assessed in detail and the cost assumptions in the WEM were revised to take account of recent cost inflation (see Chapter 6).
1. A detailed description of the WEM can be found at www.worldenergyoutlook.org.
Chapter 1 - Context and principal assumptions
61
© OECD/IEA, 2008
In addition, the integration of the WEM into a general equilibrium model, started in 2007, was taken a step further, in order to model more precisely the feedback links between energy markets and the macro-economy.
Although the Reference Scenario assumes the implementation of currently adopted energy and environmental policies during the projection period, how those policies are implemented in practice is nonetheless assumed to vary by fuel and by country. For example, while nuclear energy is assumed to remain an option for power generation in those countries that have not officially banned it or decided to phase it out (though not for others), the pace of construction of new plants differs among countries, reflecting differences in the degree of commitment to that energy source. Pricing and other market reforms are also assumed to be implemented at different rates. In most non-OECD countries, at least one fuel or form of energy continues to be subsidised, most often through price controls that hold the retail or wholesale price below the true market level. Other forms of direct financial intervention by government, such as grants, tax rebates or deductions and soft loans, are commonplace. Indirect interventions also occur, such as the free provision of energy infrastructure and services. Some of these interventions are more justified than others; for example, support to overcome market obstacles to the development and deployment of new technology. According to our calculations,2 energy-related consumption subsidies, which encourage consumption by pricing energy below market levels, amounted in 20 non-OECD countries (accounting for over 80% of total non-OECD primary energy demand) to about $310 billion in 2007 (Figure 1.1). Oil products account for one-half, or $150 billion. Most of these countries do have policies eventually to phase out consumption subsidies and we accordingly assume that these subsidies are gradually reduced, though at varying rates across regions. Consumption subsidies are minimal in OECD countries. In all cases, the rates of ad valorem taxes and excise duties on fuels are assumed to remain unchanged. Figure 1.1
Energy subsidies by fuel in non-OECD countries, 2007 Oil
Iran Russia China Saudi Arabia India Venezuela Indonesia Egypt Ukraine Argentina South Africa Kazakhstan Pakistan Malaysia Thailand Nigeria Chinese Taipei Vietnam Brazil
Gas Coal Electricity
10
20
30
40
50 60 Billion dollars
Source: IEA analysis. 2. See World Energy Outlook 2006 (Box 11.2) for a discussion of the price-gap methodology used to estimate subsidies (IEA, 2006). The IEA plans to publish a special report on subsidies in 2009.
62
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
0
1
Population Population growth affects the size and composition of energy demand, directly and through its impact on economic growth and development. Our population assumptions are drawn from the most recent United Nations projections (UNPD, 2007a). World population is projected to grow from an estimated 6.5 billion in 2006 to around 8.2 billion in 2030 — an average rate of increase of 1% per year (Figure 1.2). By 2030, 52% of the world’s population will be in non-OECD Asia, down slightly from 53% today. China will remain the world’s most heavily populated country, with more than 1.46 billion people, though India’s population all but reaches China’s by 2030. The OECD’s share of global population is set to slide further, from 18% today to less than 16% in 2030. Figure 1.2
Population by major region
Africa
2006
China
2030
India Other Asia Latin America OECD Europe OECD North America E. Europe/Eurasia Middle East OECD Pacific
0
200
400
600
800
1 000
1 200
1 400
1 600 Million
All of the increase in world population in aggregate will occur in urban areas: the rural population is expected to start to decline in about a decade. In 2008, for the first time in history, the urban population will equal the rural population of the world; from Chapter 1 - Context and principal assumptions
63
© OECD/IEA, 2008
The emerging economies will continue to experience the fastest rate of population growth. The population of non-OECD countries in total is projected to grow by 1.1% per year, against 0.4% per year in the OECD (Table 1.2). The population of all regions expands, with the exception of OECD Pacific, largely due to Japan, where population is projected to fall by 0.3% per year, and Eastern Europe/Eurasia, largely due to Russia, where population slides by 0.6% per year. By 2030, Russia’s population is expected to be 13% smaller than today. Most of the growth in the OECD population comes from North America. Several European countries, including Germany and Italy, are expected to experience significant population declines. All regions will experience continued ageing in the coming decades as fertility and mortality rates decline. Worldwide, the proportion of people over 60 years old is projected to rise from 11% now to about 15% by 2030 (UNPD, 2007b). This will have far-reaching economic and social consequences, which will inevitably affect both the level and pattern of energy use. Older people, for example, tend to travel less for work and leisure.
then on, the greater part of the world population will be urban (UNPD, 2007c). Major parts of the world, particularly in Africa, will remain largely rural. Continuing rapid urbanisation in non-OECD countries will have a major impact on demand for modern energy, the bulk of which is consumed in or close to towns and cities (see Chapter 8). Table 1.2
Population growth by region (average annual growth rates) 1980-1990
1990-2006
2006-2015
2006-2030
OECD North America United States Europe Pacific Japan
0.8% 1.2% 0.9% 0.5% 0.8% 0.6%
0.8% 1.2% 1.1% 0.5% 0.5% 0.2%
0.5% 0.9% 0.9% 0.3% 0.1% –0.1%
0.4% 0.8% 0.8% 0.2% –0.1% –0.3%
Non-OECD E. Europe/Eurasia Russia Asia China India Middle East Africa Latin America Brazil
2.0% 0.8% n.a. 1.8% 1.5% 2.1% 3.6% 2.9% 2.0% 2.1%
1.5% -0.2% -0.3% 1.4% 0.9% 1.7% 2.3% 2.4% 1.5% 1.5%
1.3% -0.2% -0.5% 1.1% 0.6% 1.4% 1.9% 2.2% 1.2% 1.2%
1.1% -0.2% -0.6% 0.9% 0.4% 1.1% 1.7% 2.0% 1.0% 0.9%
World
1.7%
1.4%
1.1%
1.0%
n.a.
0.3%
0.1%
0.0%
European Union
Note: These assumptions also apply to the climate-change scenarios described in Part C.
Economic growth is by far the most important determinant of the overall demand for energy services. The pattern of economic development also affects the fuel mix. As a result, the energy projections in the Outlook are highly sensitive to underlying assumptions about rates and patterns of economic growth. Extremely rapid economic expansion in many countries outside the OECD — especially in Asia – is the main reason why energy demand has accelerated in recent years. Nonetheless, over the long term, the income elasticity of energy demand — the change in demand relative to the change in GDP – has fallen in most parts of the world, though it has picked up since the start of the current decade. Between 1971 and 1990, each 1% increase in global GDP (in purchasing power parity, or PPP, terms) was accompanied by a 0.66% increase in primary energy consumption.3 Between 1990 and 2000, the corresponding increase in demand dropped to only 0.44%, but it rebounded to 0.68% in 2000-2006, largely due to an increase in the energy intensity of China’s economic growth. 3. All GDP data cited in this report are expressed in year-2007 dollars using purchasing power parities (PPP) (Box 1.2).
64
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Economic growth
1
There are strong signs that the global economy, which has expanded briskly in recent years, is slowing, partly as a consequence of higher energy and other commodity prices and the financial crisis that began in 2007. GDP growth has already fallen in most OECD economies, particularly in the United States, where a slump in the housing market has contributed to the credit squeeze, the consequences of which are being felt throughout the global financial system. Many OECD countries are now on the verge of recession. Most non-OECD countries have so far been less affected by financial market turbulence and have continued to grow rapidly, led by China, India and the Middle East (where growth has been boosted by surging oil-export revenues). This momentum is provided by strong productivity gains as these countries progressively integrate into the global economy, by improved terms of trade for commodity producers on the back of soaring prices for oil and other raw materials and by policy improvements. But there are signs that activity in developing Asian countries, too, is now beginning to slow. Headline inflation has increased around the world, especially in the emerging economies, with surging food and energy prices. The International Monetary Fund (IMF) projects world GDP in PPP terms to drop sharply from an above-trend 5% in 2006 and 2007 to 4.1% in 2008 and 3.9% in 2009.4 It projects inflation to rise from 2.2% in 2007 to 3.4% in 2008 in the advanced economies (essentially those of the OECD) and from 6.4% to 9.1% in the emerging economies. Trade in goods and services is expected to slow. GDP growth in all regions is expected to fall, though non-OECD countries will continue to outpace the remainder of the world. In the United States, growth is projected to fall from 2.2% in 2007 to only 1.3% in 2008 and 0.8% in 2009, while China’s growth falls from 11.9% to 9.7% in 2008 and 9.8% in 2009.
This Outlook assumes that economic growth will recover to around 4.5% per year by the turn of the decade, but will slow progressively through to 2030. World GDP is assumed to grow by an average of 3.3% per year over the period 2006-2030 compared with 3.2% from 1990 to 2006 (Table 1.3). The increase reflects the rising weight in the world economy of the non-OECD countries, where growth will remain fastest. GDP growth is assumed to average 4.2% per year in 2006-2015 and 2.8% per year in 2015-2030. Growth is nonetheless somewhat lower than assumed in last year’s Outlook, especially in OECD countries, in part because of the impact of higher energy prices and weaker prospects for global economic growth in the near term. 4. Taken from the July update of the projections in IMF (2008a), available at: www.imf.org/external/pubs/ ft/weo/2008/update/02/index.htm.
Chapter 1 - Context and principal assumptions
65
© OECD/IEA, 2008
The continued housing-market correction in the United States, similar indications in Europe and other parts of the world, and the fall-out from the financial crisis are among the most obvious uncertainties surrounding the near-term economic outlook (IMF, 2008b). Further steep declines in asset prices and an abrupt reduction in the lending capacity of the banking system could lead to a global credit crunch, with potentially severe consequences for growth. Although the IMF judges the risks to its projections to be greater on the downside, there is some potential for an unexpectedly rapid recovery among the rich industrialised economies in 2009 if conditions in financial markets improve, lowering the current high price of credit (OECD, 2008).
Table 1.3
Real GDP growth by region (average annual growth rates) 1980-1990
1990-2006
2006-2015
2006-2030
OECD North America United States Europe Pacific Japan
3.0% 3.1% 3.3% 2.4% 4.2% 3.9%
2.5% 2.9% 2.9% 2.3% 2.2% 1.3%
2.3% 2.3% 2.1% 2.3% 2.1% 1.3%
2.0% 2.2% 2.1% 1.9% 1.6% 1.2%
Non-OECD E. Europe/Eurasia Russia Asia China India Middle East Africa Latin America Brazil
2.5% 0.0% n.a. 6.7% 8.8% 5.8% 1.3% 2.4% 1.2% 1.5%
4.4% 0.0% -0.2% 7.2% 9.8% 6.1% 4.3% 3.6% 3.2% 2.7%
6.7% 5.6% 5.7% 7.9% 9.2% 7.8% 5.4% 5.8% 4.3% 4.0%
4.8% 3.7% 3.6% 5.7% 6.1% 6.4% 4.3% 4.1% 3.1% 3.0%
World
2.8%
3.2%
4.2%
3.3%
n.a.
2.1%
2.2%
1.8%
European Union
Note: The regional aggregate growth rates are calculated based on GDP expressed in purchasing-power parity terms. These assumptions also apply to the climate-policy scenarios described in Part C.
China and India are expected to continue to grow faster than other regions, followed by the Middle East. China is assumed to grow at 6.1% per year between 2006 and 2030, the second-highest rate in the world. India, which grows fastest at 6.4% per year on average, overtakes China as the fastest-growing major country in the 2020s, reflecting its more rapid population growth and its earlier stage in the development process. The growth rates of the economies of China, India and other emerging economies are expected to slow as they mature. Middle East countries also grow relatively quickly, supported by buoyant oil revenues. GDP growth is assumed to temper gradually in all three OECD regions in response to acute competition from the emerging economies and to the ageing and stabilisation or fall of the population (which will reduce the active labour force in most countries). North America is expected to remain the fastest growing OECD economy, partly due to its more rapidly expanding and relatively young population, though the rate of GDP growth is assumed to drop from an annual average of 2.9% in 1990-2006 to 2.2% per year over the Outlook period. Europe and the Pacific are expected to see the lowest GDP growth of any major region. Based on our population and GDP-growth assumptions, per-capita incomes grow most rapidly in China and India (Figure 1.3). 66
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
These GDP assumptions take account of recent major revisions to PPPs, which lower the weight of China and other developing countries in global GDP, and explain in part the reduction in average GDP growth for the world over the projection period (Box 1.2). The economies of many regions are expected to shift away from energy-intensive heavy manufacturing towards lighter industries and services, though the pace of this process, which is well advanced in the OECD and some emerging economies, varies. Industrial production continues to grow in volume terms.
Box 1.2
1
Implications of new estimates of purchasing power parity
Purchasing power parities (PPPs) measure the amount of a given currency needed to buy the same basket of goods and services, traded and non-traded, as one unit of the reference currency — in this report, the US dollar. By adjusting for differences in price levels, PPPs can, in principle, provide a more reliable indicator than market exchange rates of the true level of economic activity globally or regionally. Conversions based on market exchange rates typically underestimate the value of domestic economic activity and the output of developing countries relative to the industrialised economies. PPP rates can deviate by a large amount from the market exchange rate between two currencies. It is important to take account of the true level of economic activity in analysing the main drivers of energy demand or comparing energy intensities among countries. In 2007, the World Bank released the results of a major update of PPP exchange rates, carried out under its International Comparison Program (World Bank, 2008). As a result, the PPP differences with market exchange rates were reduced for a number of countries, most importantly China and India, thus reducing their weights in global GDP. The PPP rates were increased for some other emerging countries, including oil exporters. Overall, the average annual rate of global GDP growth between 2000 and 2007 was revised downwards by around one-half of a percentage point. China still ranks as the world’s secondlargest economy, accounting for about 11% of world GDP in 2007, compared with about 16% using the old PPP rate. The share of the United States in global GDP has been revised up from 19% to 21%. In spite of these changes, it remains true that the emerging economies have recently been the main driver of global growth in PPP terms, led by China, which contributed nearly 27% to global growth in 2007 (IMF, 2008a).
Figure 1.3
Rate of growth of per-capita income by region
OECD North America
1980-2006*
OECD Europe
2006-2030
OECD Pacific Africa Latin America Middle East Other Asia E. Europe/Eurasia
0% 1% 2% * 1990-2006 for E. Europe/Eurasia.
3%
Chapter 1 - Context and principal assumptions
4%
5%
6%
7%
8%
9%
67
© OECD/IEA, 2008
India China
Energy prices Energy prices are an exogenous determinant of the demand for and supply of each fuel and energy carrier in the WEM. The assumed trajectories for international energy prices in the Reference Scenario, summarised in Table 1.4, are based on a top-down assessment of the prices that would be needed to encourage sufficient investment in supply to meet projected demand over the Outlook period. In other words, they are derived iteratively to ensure their consistency with the overall global balance of supply and demand. They should not be interpreted as forecasts. Although the price paths follow smooth trends, this should not be taken as a prediction of stable energy markets: prices will, in reality, certainly deviate widely at times from these assumed trends. International prices are used to derive average end-user pre-tax prices for oil, gas and coal in each WEO region. Tax rates and subsidies are taken into account in calculating final post-tax prices, which help to determine final energy demand. Final electricity prices are derived from changes in marginal power-generation costs. All prices are expressed in US dollars and assume no change in exchange rates. Table 1.4
Fossil-fuel price assumptions (dollars per unit) Unit
2000
2007
2010
2015
2020
2025
2030
barrel
33.33
69.33
100.00
100.00
110.00
116.00
122.00
US imports
MBtu
4.61
6.75
12.78
13.20
14.57
15.35
16.13
European imports
MBtu
3.35
7.03
11.15
11.50
12.71
13.45
14.19
Japan LNG
MBtu
5.63
7.80
12.70
13.16
14.52
15.28
16.05
tonne
40.06
72.84
120.00
120.00
116.67
113.33
110.00
barrel
28.00
69.33
107.34
120.27
148.23
175.13
206.37
US imports
MBtu
3.87
6.75
13.72
15.88
19.64
23.18
27.28
European imports
MBtu
2.82
7.03
11.97
13.83
17.13
20.31
24.00
Japan LNG
MBtu
4.73
7.80
13.63
15.83
19.56
23.08
27.16
tonne
33.65
72.84
128.81
144.32
157.21
171.11
186.07
Real terms (2007 prices) IEA crude oil imports Natural gas
OECD steam coal imports Nominal terms IEA crude oil imports Natural gas
OECD steam coal imports
Note: Prices in the first two columns represent historical data. Gas prices are expressed on a gross calorificvalue basis. All prices are for bulk supplies exclusive of tax. Nominal prices assume inflation of 2.3% per year from 2008.
The average IEA crude oil import price, a proxy for international prices, is assumed in the Reference Scenario to average $100 per barrel in real year-2007 dollars over the period 2008-2015 and then to rise in a broadly linear manner to $122 in 2030 (Figure 1.4).5 This 5. In 2007, the average IEA crude oil import price was $2.87 per barrel lower than first-month forward West Texas Intermediate (WTI) and $3.06 lower than dated Brent.
68
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Oil prices
1
represents a major upward adjustment from WEO-2007, reflecting to a large degree higher prices for near-term physical delivery and for futures contracts out to the middle of the next decade.6 It also reflects a reassessment of the prospective marginal cost of oil supply, discussed in detail in Part B, and the outlook for demand, described in Chapter 3. In nominal terms, prices double to just over $200 per barrel in 2030.7
Dollars per barrel
Figure 1.4
Average IEA crude oil import price (annual data)
225
Real ($2007)
200
Nominal
175 150 125 100 75 50 25 0 1970
1980
1990
2000
2010
2020
2030
The surge in oil prices since the end of 2003, and especially since the beginning of 2007, can legitimately be described as an oil shock, albeit a slow-motion one. The price of West Texas Intermediate (WTI) — a leading benchmark crude — rose steadily from a low of $28 per barrel on average in September 2003 to $74 in July 2006, before falling back to $54 in January 2007. The price then resumed its upward path, hitting an intra-day peak of over $145 in July 2008 — an all-time record in both nominal and real terms. Prices rose on average in 41 of the 59 months between September 2003 and August 2008, and in all but four of the 17 months since February 2007. Prices have, however, since fallen back substantially, to around $100 in mid-September 2008.
6. Other organisations, including the US Department of Energy and the Organization of the Petroleum Exporting Countries, have also recently revised their oil-price assumptions. 7. The dollar exchange rates used were those prevailing in mid-2008 (€0.64 and ¥106), which were assumed to remain unchanged over the projection period.
Chapter 1 - Context and principal assumptions
69
© OECD/IEA, 2008
A crucial difference with the price shocks of the 1970s is that the latter were caused by supply disruptions, whereas the recent shock has resulted largely from a tightening of market conditions, as demand (particularly for middle distillates) outstripped the growth in installed crude-production and oil-refining capacity, and from growing expectations of continuing supply-side constraints in the future — a risk highlighted in the Deferred Investment Scenario of past WEOs. Geopolitical factors, including civil strife in Nigeria and worries over Iran’s nuclear programme, have magnified these upward pressures on prices. The role of speculation and the emergence of oil as an “asset class” for investors, especially in index commodity funds, remains unclear — partly because of a lack of information on real financial flows into the oil market
(see the Spotlight below).8 In contrast, there is widespread agreement among analysts that oil prices have been bid up partly to hedge against the fall in the value of the US dollar — especially since 2007. Stripping out the currency impact since 2007 shows a close fit between oil prices and global balances.9 The sudden drop in oil prices in August and early September 2008 – in the absence of any obvious major shift in demand or supply — lends support to the argument that financial investors have been playing a significant role in amplifying the impact of tighter market fundamentals on prices. Rarely has the outlook for oil prices been more uncertain than now, in mid-September 2008. We assume in our Reference Scenario that the prices reached in mid-2008 are, to some extent, the result of transient factors and that a downward correction in prices that began in August continues through to the end of the year – in part thanks to some easing of market fundamentals. Our near-term analysis of investment and new capacity additions (set out in Chapter 13) points to little change in the balance in oil supply and demand through to the middle of the next decade; that is, projected increases in installed capacity are likely to be broadly matched by rising demand in the Reference Scenario, assuming investment is sustained beyond the end of the current decade. However, pronounced short-term swings in prices are likely to remain the norm and temporary price spikes or collapses cannot be ruled out. Prices are likely to remain highly volatile, especially in the next year or two. A worsening of the current financial crisis would most likely depress economic activity and, therefore, oil demand, exerting downward pressure on prices. On the other hand, any weakening of the dollar would help support prices. Beyond 2015, we assume that rising marginal costs of supply exert upward pressure on prices through to the end of the projection period. There are acute risks to these assumptions on both sides: a continuing surge in demand, under-investment in oil-production or refining capacity, or a large and sustained supply disruption could result in higher prices. A further slide in the dollar could also push prices up in dollar terms (assuming no change in exchange rates). Yet a slump in demand — possibly caused by recession and more rapid oil-price subsidy reform than we assume — together with faster than expected growth in investment could drive prices down below our assumed level.
8. An IEA Expert Roundtable in March 2008 discussed the many influences on the current oil price. A wide range of views on the effects of money flows on the oil market was expressed, but no agreement was reached on the extent to which those flows affect prices. In the IEA’s opinion, the limited information available makes it impossible to account meaningfully for the cross-market interactions that routinely take place between different futures exchanges and over-the-counter (OTC) markets. Gaps in market as well as financial data obscure the real drivers of prices. 9. IEA, Oil Market Report, May 2008. 10. The impact of a phase-out of oil subsidies is discussed in Chapter 3.
70
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
The assumption that governments implement policies to reduce subsidies (see above) has important implications for final oil prices to end users.10 In 2007, most of the growth in world oil demand came from countries where oil-product prices are subsidised. In many countries, subsidies have risen sharply when governments have failed to increase regulated retail and wholesale prices as quickly as international prices. Iran, where retail oil prices are far below international levels, is an extreme example: the cost of those subsidies is estimated at $36 billion in 2007. Some countries, especially in Asia,
1
are now starting to rein in subsidies in the face of the mounting burden on government budgets, worries about their impact on incentives for refiners and importers to supply local markets and the prospect of persistently high prices in the medium term. Several countries, including China, India, Malaysia and Indonesia, increased prices in the first half of 2008, but some are now considering cutting them again, with the drop in international prices.
S P O T L I G H T
Are speculators to blame for soaring oil prices? As is often the case when the price of any commodity jumps, the finger of blame for the recent surge in oil prices has been pointed squarely at speculators. The amount of money invested in commodity funds has certainly risen strongly in recent years –20-fold since 2003 to more than $250 billion on some estimates. A growing number of analysts believe that this represents a speculative bubble rather than the outcome of market fundamentals. Several governments, including that of India, have called for restrictions on speculation to be imposed. But it is far from certain that the flow of money into oil trading has, in itself, driven up prices. Speculators, or financial investors, play both sides of the market, and what they buy, they eventually must sell (unless they take delivery and store physical oil, which is rare) and vice versa. Each barrel of oil that is bought for future delivery on the futures market must be sold on before the contract expires. Whatever the volume of futures trading (and there is no limit to the number of “paper” barrels that can be bought and sold), no oil is actually kept off the market. The amount of money currently invested in index funds is large in absolute terms, but pales beside the estimated $50 trillion-worth of contracts of all types traded on futures exchanges and over-the-counter spot markets in 2007.
In the end, virtually the only buyers of physical crude oil are refiners, and their needs have continued to rise with robust end-use demand, especially for middle distillates. The fact that in recent months stocks have barely risen above average levels of the past five years suggests that supply has only just managed to keep pace with demand. Thus, physical market fundamentals appear to have played the leading role in driving up prices, though financial investors may have amplified the impact.
Chapter 1 - Context and principal assumptions
71
© OECD/IEA, 2008
Indeed, more speculative money left the regulated oil market than entered it over the first half of 2008. By mid-year, the scale of net long positions (that is, a commitment to buy at some point in the future) held by non-commercial participants in futures exchanges was roughly the same as it was in July 2007, when prices were little more than half as high. Moreover, the recent tightening of credit conditions has reduced liquidity in the world’s financial system, which arguably helped fuel the speculative inflows into oil futures in previous years.
Natural gas prices Natural gas prices have remained strongly linked to oil prices, either directly through indexation clauses in long-term supply contracts or indirectly through competition between gas and oil products in power generation and end-use markets. Thus, the prices of gas traded internationally and locally on open markets around the world have risen sharply in recent years with oil prices, though price differentials can fluctuate sharply.11 Although only a small share of world gas is traded between regions, the links between oil and gas markets mean that regional gas prices tend to move in parallel. Lags in long-term contracts, which still account for a large share of the gas traded outside North America, mean that gas prices have not yet fully reflected the increase in oil prices. That is the main reason why gas prices have generally risen less rapidly than oil prices since the beginning of the current decade (Figure 1.5). Gas prices on average in early 2008 were highest in North America, where most gas is traded on short-term markets and few lags exist, but prices had fallen back sharply by mid-year, reflecting a surge in production, weak demand and rising stocks. In our Reference Scenario, gas prices jump in all three regions in 2008 and then fall back slightly through to 2010 in lagged response to the fall in oil prices from their mid2008 peaks. Prices begin to rise after 2015 in line with oil prices. Although the gradual development of gas-to-gas competition is expected to weaken the contractual links between oil and gas prices over the projection period, gas prices are not expected to fall relative to oil prices. Competition would be expected to exert some downward pressure on the prices of gas relative to those of oil. This is assumed to be offset by rising marginal supply costs for gas as the distances over which gas has to be transported, by pipeline or as liquefied natural gas (LNG) increase. The growing share of LNG in global gas supply and increasing opportunities for short-term trading of LNG are expected to contribute to a modest degree of convergence in regional prices over the projection period.
Fuel price divided by oil price
Figure 1.5
Assumed natural gas and coal prices relative to crude oil*
1.6
Gas - Europe
1.4
Gas - Japan
1.2
Gas - United States
1.0
Coal - OECD
0.8 0.6 0.4 0.2 0 1980
1990
2000
2010
2020
2030
11. The lags have nonetheless shortened, as more recent long-term contracts have generally been based on oil prices in the previous 3 to 6 months, rather than 6 to 12 months as in older contracts.
72
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
* Calculated on an energy-equivalent basis using real-2007 dollars.
1
Steam coal prices International steam coal prices have surged in recent years. The average price of coal imported by OECD countries jumped from $42 per tonne in 2003 to $73 in 2007 (in year2007 dollars) and soared to well over $100 in the first half of 2008.12 Rising industrial production and electricity demand, especially in China, have boosted coal use. Higher gas prices have also encouraged some power stations and industrial end users to switch to coal and to invest in new coal-fired equipment. These factors have added to the upward pressure on coal demand and prices. Coal prices are assumed to settle at around $120 per tonne in real terms in 2010. Thereafter, prices are assumed to remain flat through to 2015 and then to fall back slightly to $110 in 2030 as new mining and transportation capacity becomes available. As oil and gas prices are assumed to rise steadily after 2015, coal becomes increasingly competitive – at least in countries not actively seeking to limit CO2 emissions. In reality, the possibility of a carbon value being introduced or increasing where it already exists, together with a tightening of other environmental regulations, is likely to affect the use of coal in all regions, or to increase the cost of using it, counterbalancing the impact on coal demand of relatively lower prices.
Technology Technological innovation and the rate of deployment of new technologies for supplying or using energy have a major impact on energy balances, both in terms of the overall amount of energy used and the fuel mix. Our projections are, therefore, very sensitive to assumptions about technological developments. In general, it is assumed in the Reference Scenario that the performance of currently available technologies improves on various operational criteria, including energy efficiency. High energy prices are expected to contribute to this. However, our assumptions about the pace of technological advance vary for each fuel and each sector, depending on our assessment of the potential for efficiency improvements and the stage of technology development and commercialisation. Crucially, no altogether new technologies on the demand or supply side, beyond those known about today, are assumed to be deployed before the end of the projection period, since it cannot be known whether or when such breakthroughs might occur and how quickly they may be commercialised.
12. In mid-2008, prices approached $200 per tonne for certain qualities of coal in some European markets and $150 in the United States.
Chapter 1 - Context and principal assumptions
73
© OECD/IEA, 2008
The average energy efficiency of equipment currently in use and the overall intensity of energy consumption are typically assumed to improve progressively over the projection period. However, the rate of improvement varies considerably among different types of equipment, primarily according to the rate of retirement and replacement of the capital stock. Since much of the energy-using capital stock in use today will be replaced only gradually, most of the impact of recent and future technological developments that improve energy efficiency will not be felt until near the end of the projection
period. Rates of capital-stock turnover differ greatly (Figure 1.6). Most of today’s cars and trucks, heating and cooling systems, and industrial boilers will be replaced by 2030. But most existing buildings, roads, railways and airports, as well as many power stations and refineries, will still be in use then, barring strong government action to encourage or force early retirement (which is not assumed in the Reference Scenario). Despite slow capital stock turnover, refurbishment could, in some cases, significantly improve energy efficiency at acceptable economic cost. On the supply side, technological advances are assumed to improve the technical and economic efficiency of producing and supplying energy. In some cases, they are expected to reduce unit costs and to lead to new and cleaner ways of producing and delivering energy services. For example, power-generation thermal efficiencies – as measured by the amount of useful electricity and heat energy produced, divided by the amount of energy contained in the fuel input – are assumed to improve over the projection period, but at different rates for different technologies. Exploration and production techniques for oil and gas are also expected to improve, boosting exploration success rates, raising recovery rates and opening up new opportunities for developing hard-to-exploit resources. Upstream technology research and development spending by the oil and gas industry has been rising rapidly in recent years (see Part B). Some major new supply-side technologies that are approaching the commercialisation phase are assumed to become available and to be deployed to some degree before the end of the projection period. These include:
Second-generation biofuels: New biofuels technologies being developed today, notably hydrolysis and gasification of woody ligno-cellulosic feedstock to produce ethanol, are assumed to reach commercialisation by around 2020. Although the technology already exists, more research is needed to improve process efficiencies, especially enzymatic hydrolysis to extract fermentable sugar from the lignocellulosic material contained in plant material and to produce a fermented broth with a higher concentration of ethanol. There is virtually no commercial production 74
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Carbon capture and storage (CCS): This is a promising option for mitigating emissions of CO2 from power plants and other industrial facilities, though it has not yet been deployed on a significant scale (IEA, 2008). The basic technology already exists to capture the gas and to transport and store it permanently in geological formations. There are at present four large-scale CCS projects in operation around the world, each involving the separation of around 1 Mt of CO2 per year from produced natural gas: Sleipner and Snohvit in Norway, Weyburn in Canada (with the CO2 sourced in the United States) and In Salah in Algeria. The captured CO2 is used for enhanced oil recovery in the latter two projects. Yet there are a number of technical, economic and legal barriers to the more widespread and rapid deployment of CCS. In particular, it is very energy-intensive and expensive. Considerable research and development effort is currently going into CCS in power generation. Around 20 demonstration projects are already under construction or are planned, which are expected to lead to lower unit costs and improved operational performance. We assume that CCS will become commercially available after 2020 in countries where adequate financial incentives and/or regulatory measures to reduce CO2 emissions are already in place.
Figure 1.6
1
Typical lifetimes of energy-related capital stock Land use and housing
Urban infrastructure*
Power generation Building stock*
Transformation, transport and distribution
Large hydropower plant
Final consumption
Coal-fired power plant Nuclear power plant Combined-cycle gas-turbine plant Fuel cells (power generation) Photovoltaics Wind turbines Electricity transmission & distribution Pipelines Oil refineries Refueling stations Aircraft desgin (from first studies to series-end) Cement plant Aircraft (operating lifetime) Petrochemical plant Steel plant Manufacturing equipment Residential space heating and cooling Commercial heating and cooling equipment Passenger cars Household appliances (stoves, washers, etc.) Residential water heating equipment Consumer electronics Light bulbs (fluorescent) Office equipment
0
20
40
60
80
100
* The upper range exceeds 140 years. Note: Bars indicate average lifetimes, lines indicate the typical range. Source: IEA analysis.
Chapter 1 - Context and principal assumptions
120 140 Years
75
© OECD/IEA, 2008
Light bulbs (incandescent)
of ethanol yet from cellulosic biomass, but there is substantial research going on in this area in several OECD countries. The US Energy Independence and Security Act mandated investment in advanced biofuels. Commercial-scale demonstration plants are under construction in the United States and several others are planned in Europe and other parts of the world.
76
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Coal-to-liquids (CTL): The indirect conversion of coal to oil products through gasification and Fischer-Tropsch synthesis processes – much like gas-to-liquids production – has been carried out commercially for many decades. Yet global production remains limited as CTL production has in most cases been uneconomic up to now, mainly because of the large amounts of energy and water used in the process, the high cost of building CTL plants and volatile oil and coal prices. Recently, research has been focused on direct liquefaction, which is much less energy-intensive than the indirect CTL method, though the liquids produced require further refining to meet applicable fuel standards, increasing unit costs. The Chinese coal company Shenhua is commissioning in 2008 a 20 000-b/d demonstration plant at Erdos in Inner Mongolia and more plants are planned in China and the United States. The full commercialisation of direct liquefaction technologies is expected to develop over the Outlook period, boosting their deployment in countries with abundant low-cost coal and freshwater resources (see Chapter 11 for detailed projections of CTL production).
CHAPTER 2
GLOBAL ENERGY TRENDS H
I
G
H
L
I
G
H
T
S
World energy use continues to increase steadily in the Reference Scenario, but at a slower rate than projected in WEO-2007, mainly due to higher energy prices and slower economic growth. Global primary energy demand grows by 1.6% per year on average in 2006-2030. Oil demand increases progressively, though more slowly than in WEO-2007, particularly in the second half of the Outlook period. Fossil fuels account for 80% of the world’s primary energy mix in 2030 — down slightly on today. Oil remains the dominant fuel, though demand for coal rises more than demand for any other fuel in absolute terms. The share of natural gas in total energy demand rises marginally, with most of the growth coming from the power-generation sector. Coal continues to account for about half of fuel needs for power generation. The contribution of non-hydro renewables to meeting primary energy needs inches up from 11% now to 12% in 2030. Due to strong economic growth, China and India account for 51% of incremental world primary energy demand in 2006-2030. Middle East countries emerge as an important demand centre. Of the global increase in oil demand, 43% comes from China, 20% from the Middle East and 19% from India. Over a quarter of the growth in world gas demand comes from the Middle East. Non-OECD countries account for 87% of the increase in global demand between 2006 and 2030. As a result, their share of world primary energy demand rises from 51% to 62%. Industry overtakes transport before 2010 to become the second-largest final energy-consuming sector, after the combined residential, services and agricultural sector. Among all final energy forms, electricity consumption grows fastest, nearly doubling in 2006-2030, boosting its share in total final energy consumption from 17% to 21%. Oil remains the single largest end-use fuel, though its share drops from 43% in 2006 to 40% in 2030.
Cumulative investment needs amount to $26.3 trillion (in year-2007 dollars) in 2007-2030, over $4 trillion more than posited in WEO-2007. The power sector accounts for $13.6 trillion, or 52% of the total. To provide adequate assurance for future investment in energy-supply infrastructure, negotiations need to be concluded on an international agreement on combating climate change and the implications for national policies quickly assessed.
Chapter 2 - Global energy trends
77
© OECD/IEA, 2008
Almost all of the increase in fossil-fuel production over the Outlook period occurs in non-OECD countries. The Middle East and Africa are the biggest contributors to increased exports. As a result, the reliance on imported oil and gas of the main consuming regions, including the OECD and Asian economies, increases substantially, particularly in the second half of the projection period.
Demand Primary energy mix Projected world primary energy demand in the Reference Scenario increases by 45% between 2006 and 2030 – an average annual rate of growth of 1.6% (Table 2.1).1 This is 0.2 percentage points slower than the rate projected in last year’s Outlook, largely due to changes in our assumptions for energy prices and economic growth, as well as new government policies introduced since last year (Box 2.1). It is also slower than the average growth of 1.9% per year from 1980 to 2006. World energy intensity — primary energy demand per unit of real GDP (in purchasing power parity, or PPP, terms) — is expected to decline over the projection period by 1.7% a year, 0.6 percentage points faster than over the past three decades. Lower energy intensity is primarily the result of acceleration in the transition to a service economy in many major non-OECD countries and more rapid efficiency improvements in the power and end-use sectors in the OECD. Table 2.1
World primary energy demand by fuel in the Reference Scenario (Mtoe) 1980
2000
2006
2015
2030
20062030*
Coal
1 788
2 295
3 053
4 023
4 908
2.0%
Oil
3 107
3 649
4 029
4 525
5 109
1.0%
Gas
1 235
2 088
2 407
2 903
3 670
1.8%
186
675
728
817
901
0.9%
Nuclear Hydro
148
225
261
321
414
1.9%
Biomass and waste**
748
1 045
1 186
1 375
1 662
1.4%
Other renewables Total
12
55
66
158
350
7.2%
7 223
10 034
11 730
14 121
17 014
1.6%
World demand for coal advances by 2% a year on average, its share in global energy demand climbing from 26% in 2006 to 29% in 2030. Most of the increase in demand for coal comes from the power-generation sector. China and India together contribute 85% to the increase in world coal demand over the projection period. Oil remains the dominant fuel in the primary energy mix, but its share drops to 30% in 2030, from 34% in 2006. Oil demand grows far more slowly than demand for other fossil fuels, mainly because of high final prices. Gas demand increases at 1.8% per annum over the projection period and its share in world primary energy moves up slightly, to 22% in 2030. New power plants, using high-efficiency gas turbine technology for the most part, meet the bulk of incremental gas demand. The Middle East, developing Asian countries and the OECD see the biggest increases in gas demand. The share of nuclear power in primary energy demand edges down over the Outlook period, from 6% today to 5% in 2030, reflecting the assumption of unchanging national 1. Projections of energy-related CO2 emissions and local pollutants are covered in Chapter 16.
78
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
* Average annual rate of growth. ** Includes traditional and modern uses.
policies towards nuclear power. Nuclear output nonetheless increases in absolute terms in all major regions except OECD Europe. The largest increases will take place in developing Asia. There is considerable uncertainty about the prospects for nuclear power, given a recent revival of interest in re-evaluating the role of nuclear power in order to combat climate change. A change in policy could lead to a significantly higher share of nuclear power in the energy mix than assumed here (see Chapter 6).
Box 2.1
2
How do the new WEO assumptions affect global energy trends compared with last year?
Higher energy prices and slower economic growth than assumed last year are the main reasons why demand grows more slowly in this year’s Outlook. World GDP is assumed to grow on average by 3.3% per year in 2006-2030, compared with 3.6% in WEO-2007. The IEA crude oil import price reaches $122 per barrel in real terms in 2030 in this year’s WEO, very nearly twice the level in WEO-2007. Lower economic growth reduces the overall demand for energy services. Higher prices encourage consumers and businesses to conserve energy and to buy more efficient appliances and equipment. They also encourage manufacturers to develop more energy-efficient equipment.
As a result of these changes, demand for each fossil fuel grows less rapidly than projected last year. Global demand for coal is 86 Mtoe lower in 2030 than in WEO-2007, while gas demand is 278 Mtoe lower. The biggest change, however, is in global oil demand. In 2030, world primary oil consumption is 5 109 Mtoe in this year’s Outlook, compared with 5 585 Mtoe last year. Higher oil prices and vehicle fuel-efficiency improvements are the main drivers of this change. Faster growth in the use of biofuels for transport, aided by the penetration of secondgeneration biofuel technologies towards the end of the Outlook period, also contributes. Excluding hydropower and biomass and waste, other renewables grow more quickly in this WEO, by 284 Mtoe in 2006-2030, compared with 247 Mtoe in 2005-2030 in last year’s WEO.
Chapter 2 - Global energy trends
79
© OECD/IEA, 2008
Assumptions about policy and technology have also changed. A number of countries, mostly in the OECD, have adopted during the past year new policies that substantially affect energy-demand prospects. These are taken into account in the Reference Scenario. One example is the US Energy Independence and Security Act, which mandates vehicle fuel-economy improvements and higher production of biofuels (see Chapter 1). This year’s WEO also assumes that carbon capture and storage (CCS) will be introduced in the power-generation and industry sectors towards the end of the projection period, although only a few plants are assumed to be in operation in 2030 (nearly all of them in OECD countries and China). By 2030, coal plants with CCS consume about 0.2% of all the coal used worldwide. This year’s supply-side projections also reflect higher marginal costs and faster decline rate assumptions in the oil and gas sector in key producing countries.
Hydropower has long been a major source of electricity production and its relative importance does not change in the Reference Scenario. While most of the OECD’s lowcost, hydro-electric resources have already been exploited, several large-scale projects in non-OECD countries are expected to be launched over the Outlook period (see Chapter 7). World hydropower production grows by an average 1.9% a year, its share of primary demand remaining constant at 2%. Hydropower’s share in global electricity generation, however, drops two percentage points to 14% in 2030.
Mtoe
Figure 2.1
World primary energy demand by fuel in the Reference Scenario
6 000
Oil Coal
5 000
Gas 4 000
Biomass
3 000
Nuclear Hydro
2 000
Other renewables
1 000 0 1980
1990
2000
2010
2020
2030
The use of biomass and waste for energy increases by 1.4% per year.2 This figure masks considerable differences among countries in how this energy source is used: the use of biomass in modern applications such as biofuels and power generation rises quickly, while the use of traditional biomass in inefficient cooking stoves in poor households in less developed parts of the world grows at a much slower pace. The use of biomass and waste for power generation, mainly in OECD countries, grows at 5.4% per year, albeit from a low base. Other renewables, a category that includes wind, solar, geothermal, tidal and wave energy, grow faster than any other energy source, at an average rate of 7.2% per year over the projection period. Most of the increase is in the power sector. The share of other renewables in total power generation grows from 1% in 2006 to 4% in 2030.
Energy demand in non-OECD countries exceeded that in OECD countries in 2005 for the first time ever (Figure 2.2). The faster pace of demand growth outside the OECD is set to continue. Driven mainly by brisk growth in China and India, non-OECD countries account for 87% of the increase in global demand between 2006 and 2030. As a result, the non-OECD share of world primary energy demand rises from 51% in 2006 to 62% by 2. See Chapter 7 for a discussion of traditional biomass.
80
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Regional trends
2030. Steady economic growth and industrial expansion, population increase and higher urbanisation rates drive demand growth in non-OECD countries. The replacement of fuelwood and charcoal with oil and gas also plays a major role. Growth in energy demand is fastest in the Middle East and Asia (Table 2.2).
Mtoe
Figure 2.2
2
World primary energy demand by region in the Reference Scenario
12 000
China and India
10 000
Rest of non-OECD OECD
8 000 6 000 4 000 2 000
Table 2.2
1990
2000
2010
2020
2030
World primary energy demand by region in the Reference Scenario (Mtoe) 1980
2000
2006
2015
2030
2006-2030*
OECD
4 072
5 325
5 536
5 854
6 180
0.5%
North America
2 100
2 705
2 768
2 914
3 180
0.6%
United States
1 809
2 300
2 319
2 396
2 566
0.4%
Europe
1 504
1 775
1 884
1 980
2 005
0.3%
Pacific
467
845
884
960
995
0.5%
Non-OECD
3 043
4 563
6 011
8 067
10 604
2.4%
E. Europe/Eurasia
1 267
1 015
1 118
1 317
1 454
1.1%
Russia
n.a.
615
668
798
859
1.1%
1 072
2 191
3 227
4 598
6 325
2.8%
China
604
1 122
1 898
2 906
3 885
3.0%
India
209
460
566
771
1 280
3.5%
133
389
522
760
1 106
3.2%
Asia
Middle East Africa
278
507
614
721
857
1.4%
Latin America
294
460
530
671
862
2.0%
7 223
10 034
11 730
14 121
17 014
1.6%
n.a.
1 722
1 821
1 897
1 903
0.2%
World** European Union
* Average annual rate of growth. ** World includes international marine bunkers.
Chapter 2 - Global energy trends
81
© OECD/IEA, 2008
0 1980
The volumetric increase in China’s energy demand in 2006-2030 dwarfs that of all other countries and regions, the result of its rapid economic and population growth. The almost 2 000-Mtoe increase in demand in 2006-2030 is nearly four times bigger than the combined increase in all of the countries in Latin America and Africa, and more than three times as large as the increase in the OECD (Figure 2.3). China’s contribution to incremental global oil demand is 43%. India accounts for 19%. The countries of the Middle East, most of them key oil and/or gas producers, emerge as a major oil- and gas-consuming region. It sees the second-largest increase in global oil demand, after China. This increase in the Middle East is underpinned by rapid economic growth and by persistent subsidies on oil products.3 In the case of coal, China accounts for 66% of the global increase in demand. By 2030, OECD countries account for less than a quarter of global coal use. In contrast, OECD countries account for 46% of the global increase in the use of renewables. Figure 2.3
Incremental primary energy demand by fuel in the Reference Scenario, 2006-2030 Coal
Africa
Oil
Latin America
Gas
E. Europe/Eurasia
Nuclear
Other Asia
Hydro Middle East
Other*
OECD India China –500
0
500
1 000
1 500
2 000 Mtoe
Changes in the fuel mix vary considerably across regions. In the OECD, where overall demand grows very slowly over the projection period, oil demand falls slightly, while gas and non-hydro renewables make up most of the increase in energy demand. Demand in Eastern Europe and Eurasian countries (including Russia) grows by only 336 Mtoe between 2006 and 2030, with nearly 60% of this increase taking the form of natural gas and oil. While China and India continue to rely predominately on coal, incremental energy demand in other Asian countries and in Latin America is remarkably diverse. Despite having the highest regional rate of increase in coal consumption, the Middle East countries rely almost entirely on oil and gas to meet additional energy needs over the projection period. Incremental gas demand in this region is equal to 26% of the world total. The Middle East becomes the third-largest gas consumer, after OECD North America and OECD Europe, surpassing projected demand in the entire European 3. See Chapter 1 for a detailed discussion of subsidies.
82
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
* Other includes biomass and waste, and other renewables.
Union in 2030. Under the assumptions of the Reference Scenario, the only region where nuclear power expands significantly is Asia, with most of the increase occurring in China. Biomass and waste and other renewables account for nearly 40% of incremental energy demand in Africa, consumed mostly as fuelwood, crop residues and charcoal for cooking and heating. By contrast, the 61% contribution of biomass and other non-hydro renewables to incremental energy demand in OECD countries is in the form of modern technologies, mostly wind.
2
In 2030, disparities in per-capita energy consumption among regions remain stark (Figure 2.4). Middle East countries see a rapid increase in per-capita consumption, overtaking OECD Europe by 2030. Despite the relatively small increase in energy demand over the Outlook period, Russia still has the highest per-capita energy consumption, at 7.0 toe, in 2030. Per-capita consumption increases rapidly from 1.4 toe in 2006 to 2.7 toe in 2030 in China, thanks to a booming economy and slower population growth compared with African and other Asian countries. India’s per-capita energy use is only 0.9 toe in 2030, but up from 0.5 toe in 2006. On average across countries, per-capita consumption in sub-Saharan Africa is only 0.5 toe in 2030 — about a third of the level in Latin America and one-ninth of that in OECD countries. Figure 2.4
Per-capita primary energy demand by region, 2030
Toe per capita 7 6 5 4 3 2 1 0 The boundaries and names shown and the designations used on maps included in this publication do not imply official endorsement or acceptance by the IEA.
The combined power-generation and heat sector absorbs a growing share of global primary energy demand over the projection period. Its share reaches over 42% in 2030, compared with 38% in 2006. Coal remains the leading input for powergeneration and heat, its share of total inputs holding steady at about 47% over the Outlook period. Oil’s share falls from 6% in 2006 to 3% in 2030, while the share of gas Chapter 2 - Global energy trends
83
© OECD/IEA, 2008
Sectoral trends
rises from 21% to 23%. Nuclear power’s contribution falls from 16% in 2006 to 13% in 2030. Hydropower’s share remains steady at 6%. Inputs from non-hydro renewables, mostly wind and biomass and waste, grow worldwide at an average rate of 6.2% per year between 2006 and 2030, the fastest rate of all energy sources, with their share rising to 8% in 2030. Aggregate demand in final-use sectors (industry, transport, residential, services, agriculture and non-energy use) is projected to grow by 1.4% per year from 2006 to 2030 — slightly more slowly than primary energy demand (Table 2.3). Industry demand grows most rapidly, at 1.8% per annum, overtaking transport before 2010 as the secondlargest final-use sector, after the combined residential, services and agriculture sector. Industry demand increases everywhere, but fastest in the Middle East and non-OECD Asia. The rate of growth in global transport energy demand slows considerably over the Outlook period. It averages 1.5% per year on average in 2006-2030 compared with 2.3% in 1980-2006, largely as a result of improved fuel efficiency of the vehicle fleet (see Chapter 3). Residential, services and agriculture consumption grows at an average annual rate of 1.2%, slower than the rate of growth of 1.5% per year in 1980-2006, due to efficiency improvements and fuel switching. World final energy consumption by sector in the Reference Scenario (Mtoe)
Industry
1980
2000
2006
2015
2030
20062030*
1 779
1 879
2 181
2 735
3 322
1.8%
Coal
421
405
550
713
838
1.8%
Oil
474
325
329
366
385
0.7%
Gas
422
422
434
508
604
1.4%
Electricity
297
455
560
789
1 060
2.7%
Other
165
272
307
359
436
1.5%
Transport
1 245
1 936
2 227
2 637
3 171
1.5%
Oil
1 187
1 844
2 105
2 450
2 915
1.4%
2
10
24
74
118
6.8%
57
82
98
113
137
1.4%
2 006
2 635
2 937
3 310
3 918
1.2%
Coal
244
108
114
118
100
-0.5%
Oil
481
462
472
493
560
0.7%
Gas
346
542
592
660
791
1.2%
Electricity
273
613
764
967
1 322
2.3%
Biofuels Other Residential, services and agriculture
Other
661
910
995
1 073
1 144
0.6%
Non-energy use
348
598
740
876
994
1.2%
5 378
7 048
8 086
9 560
11 405
1.4%
Total * Average annual rate of growth.
84
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Table 2.3
Among all end-use sources of energy except other renewables, electricity is projected to grow most rapidly worldwide, by 2.5% per year from 2006 to 2030.4 Electricity consumption nearly doubles over that period, pushing up its share in total final energy consumption from 17% to 21%. Electricity use expands most rapidly in non-OECD countries, by 3.8% per year; it increases by 1.1% per year in the OECD.
2
The share of coal in world final consumption remains broadly constant at about 9%. Coal use expands in industry, but only in non-OECD countries. The share of gas in world final consumption does not change. Final gas demand rises by 1.3% per year on average, with most of the increase in non-OECD countries. Oil demand grows by 1.2% per year, with almost three-quarters of the increase coming from transport. Biofuels for transport, nearly all of which are produced from first-generation technologies, account for 4% of total transport energy demand in 2030, a significant increase over the 1% share in 2006. The use of biomass and waste rises in absolute terms in the residential, services and agriculture sector, but its share of total final energy use in these sectors combined drops from 28% in 2006 to 22% in 2030. The share of electricity in this sector rises to 34% in 2030 from 26% in 2006. Despite rapid growth in some countries, per-capita electricity consumption, particularly in Africa, remains far below that of OECD countries in 2030 (see Chapter 6). Per-capita annual electricity demand in the residential sector in OECD countries averages 2 835 kWh in 2030, compared with 565 kWh in non-OECD countries. In sub-Saharan Africa, it is only 132 kWh per capita in 2030.5 Other renewables, mostly photovoltaic, achieve some penetration in the combined residential, services and agriculture sectors, but still meet only about 2% of this sector’s global energy needs in 2030.
Energy production and trade Resources and production prospects
4. Other renewables, mostly solar heating and cooling, grow from a very low base and account for only 1% of global final energy demand in 2030. 5. See Chapter 15 for a discussion of electricity access in oil- and gas-rich countries of sub-Saharan Africa.
Chapter 2 - Global energy trends
85
© OECD/IEA, 2008
We judge that the world’s overall energy resources are adequate to meet the projected growth in energy demand to 2030. The key to tapping the available resources is mobilising investment in a timely manner. Restrictions on oil and gas investment in many of the most prospective areas mean that investment does not always go to developing the cheapest reserves. Increasingly, fossil-energy supplies will have to come from deposits that are more difficult and costly to exploit, including deepwater and non-conventional oil and gas resources. Yet technologies for extracting such resources are constantly improving. We expect non-conventional and frontier resources to carve out a growing share of global oil and gas supplies by 2030 (see Part B). The related costs
will undoubtedly be higher than in the past, with the technical challenge enhanced by higher raw material and equipment costs across the board. This is one reason why we are assuming higher prices in this year’s Outlook. Proven and probable reserves of natural gas are much larger than for oil. However, as for oil, some factors — such as geopolitical and technical risks, investment restrictions and national policy constraints in resource-rich countries — cast doubt over the extent to which the long-term potential can be converted to actual supply capacity. Concerns about the potential to expand coal supplies are less pronounced, though the associated costs are uncertain. There is a large quantity of uranium resources that could be mined for nuclear power production. Although location-specific, renewable energy sources are also abundant. Most of the increase in oil production over the projection period is expected to come from a small number of countries where remaining resources are concentrated. These include several OPEC countries, mainly in the Middle East, and a handful of non-OPEC countries, notably Canada (with vast oil-sands reserves), the Caspian countries and Brazil. The majority of oil-producing countries will see their output drop. The OPEC share in total production in the Reference Scenario rises from 44% today to 51% in 2030, on the assumption that the requisite investment is forthcoming, which over the past decade has not always been the case. Natural gas liquids and non-conventional sources — mainly Canadian oil sands — make a growing contribution to world oil supplies over the Outlook period. Production of natural gas resources, which are more widely dispersed than oil, increases in all regions except OECD Europe (see Chapter 4). The biggest increases are projected to occur in the Middle East and Africa, where large, low-cost reserves are found. Gas production triples in the Middle East and more than doubles in Africa. Supply costs have been rising quickly as a result of both the cost inflation that has afflicted all parts of the energy sector and the increased length of supply chains as consuming regions become more reliant on more distant sources.
The shift in the geographical sources of incremental energy supplies over the Outlook period is pronounced. Gas production in the OECD as a whole declines by 31 bcm, due to a 88-bcm decline in production in OECD Europe. Oil production increases in the OECD, but only by 1.4 mb/d (Figure 2.5). The increase in oil production in nonOECD countries in 2007-2030 is 20 mb/d. Most of the incremental coal production comes from non-OECD countries; OECD countries contribute about 10% to the global increase. 86
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Although there are abundant coal reserves in most regions, increases in coal production are likely to be concentrated where extraction, processing and transportation costs are lowest. China is set to reinforce its position as the world’s leading coal producer, accounting for close to two-thirds of the increase in global output over the projection period in the Reference Scenario (though this is not sufficient to meet domestic demand growth). The United States, India and Australia remain the next biggest producers. Most of the increase in world coal production takes the form of steam coal.
Incremental world fossil-fuel production in the Reference Scenario 1 600
Mtce
25
Bcm
mb/d
Figure 2.5
1 400 20
3 000
2 Non-OECD OECD
2 500
1 200 2 000
1 000
15
1 500
800 10
600
1 000
400 5
500
200 0 –200 1980-2007 2007-2030 Oil
1980-2006 2006-2030 Gas
1980-2006 2006-2030 Coal
Inter-regional trade
Growing fossil-energy trade has major implications for energy security, a point highlighted in last year’s Outlook (IEA, 2007). Of the net oil-importing regions today — the three OECD regions and non-OECD Asia — OECD Europe and Asia become even more dependent on imports over the projection period (see Chapter 3). While still remaining significant oil importers, the OECD North America and Pacific regions see their net imports decline. The Middle East, already the biggest exporting region, sees its share of inter-regional oil trade rise from 49% in 2007 to 52% in 2030. Such a development intensifies concerns about the world’s vulnerability to a price shock induced by a supply disruption. Maintaining the security of international sea lanes and pipelines will become more important as oil and gas supply chains lengthen (see Figure 3.11 in Chapter 3 and Figure 4.6 in Chapter 4). Chapter 2 - Global energy trends
87
© OECD/IEA, 2008
International trade in energy is expected to expand, in both absolute terms and as a share of production, to accommodate the mismatch between the location of demand and that of production. The share of oil traded between WEO regions is projected to grow by 35% between 2007 and 2030, as production becomes increasingly concentrated in a small number of resource-rich countries. The decline in gas production in OECD countries leads to more imports in the form of liquefied natural gas (LNG) and, to a lesser extent, by pipeline. Coal trade also increases. Even China, the world’s largest coal producer, sees its coal imports rise to meet its galloping demand (see Chapter 5). Inter-regional trade in other fuels and carriers remains small.
Energy investment Trends by region and energy source The Reference Scenario projections call for cumulative investment in energy-supply infrastructure of $26.3 trillion (in year-2007 dollars) over 2007-2030 (Table 2.4). This amount is around $4.4 trillion higher than in WEO-2007, because of an upward revision in assumed unit costs. More than half of global investment, or $13.6 trillion, goes towards the power sector. Oil- and gas-sector investments total $11.7 trillion, $2.2 trillion higher than last year’s Outlook, with over 65% required in non-OECD countries. Coal-industry investments (not including transportation) are much smaller, totalling less than $730 billion, or 3% of total energy investment. The production of coal is inherently much less capital intensive than that of oil, gas or electricity. Unit capital costs, especially in the oil and gas industry, have surged in the last year, leading to an upward revision in our assumed costs for the projection period (see Chapter 13). Huge inflows of capital are needed to expand supply capacity to meet rising demand, as well as to replace existing and future supply facilities that will be retired during the projection period. Just over half of projected global energy investment goes simply to maintain the current level of supply capacity: much of the world’s current production capacity for oil, gas, coal and electricity will need to be replaced by 2030. In addition, some of the new production capacity brought on stream in the early years of the projection period will need to be replaced before 2030. Many power plants, electricity and gas transmission and distribution facilities, and oil refineries will also need to be replaced or refurbished. Cumulative investment in energy-supply infrastructure in the Reference Scenario, 2007-2030 ($ billion in year-2007 dollars)
OECD
Coal
Oil
Gas
Power
Total
165
1 437
2 286
5 708
9 739
North America
87
1 023
1 675
2 645
5 490
Europe
39
304
417
2 259
3 099
Pacific
39
110
195
804
1 149
521
4 635
3 044
7 897
16 187
53
1 079
859
916
2 913
Non-OECD E. Europe/Eurasia Russia
36
544
653
440
1 674
431
916
682
5 327
7 386
China
323
515
234
3 099
4 186
India
70
179
82
1 455
1 791
1
997
597
509
2 107
Africa
23
868
608
447
1 949
Latin America
13
775
298
697
1 832
Inter-regional transport
42
225
122
n.a.
389
728
6 296
5 452
13 604
26 315
Asia
Middle East
World
Note: Regional totals include a total of $234 billion investment in biofuels.
88
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Table 2.4
The Reference Scenario projections of electricity supply call for cumulative investment of $13.6 trillion, 58% of which is needed in non-OECD countries. Power generation accounts for half of this amount (Figure 2.6). Global needs in this sector are over $2 trillion more than projected in WEO-2007. Much of the increase is due to rising raw material and equipment costs (see Chapter 6). Financing the building of new energy infrastructure will be a major challenge, especially for poorer countries that rely on public sources of finance.
2
Due to the world’s rising reliance on oil and gas from the Middle East, Africa and other non-OECD countries, the rapid development of power infrastructure in many of these countries and fast-rising energy needs in China and India, most of the required investment is needed in non-OECD countries. Mobilising this investment will require the lowering of regulatory and market barriers. Many major oil and gas producers in Africa, the Middle East and Latin America recognise the need for — and value of — foreign involvement. Algeria, Egypt, Libya, Nigeria and several countries in the Middle East have attracted joint-venture investment by international oil companies, although the dominance of national oil companies is growing in many other countries (see Chapter 14). Key coal producers also need to attract capital to meet their mediumterm production targets. Foreign investment could bring better management and a greater willingness to use the best available technologies, as well as access to capital, to many emerging economies. Many countries are liberalising and restructuring their electricity industries in order to attract private domestic and foreign investment, and to improve the way energy companies are run.
Figure 2.6
Cumulative investment in energy-supply infrastructure in the Reference Scenario, 2007-2030 Total investment = $26.3 trillion (in year-2007 dollars) Coal 3% Biofuels 1% $0.7 trillion $0.2 trillion
Power
Oil 24% $6.3 trillion
Oil Shipping 4%
Transmission and distribution 50%
Power generation 50%
Refining 16%
Chapter 2 - Global energy trends
Exploration and development 80%
Gas 21% $5.5 trillion
Gas Transmission and Exploration distribution and 31% development LNG chain 61% 8%
Coal Shipping and ports 9%
Mining 91%
89
© OECD/IEA, 2008
Power 52% $13.6 trillion
© OECD/IEA, 2008
CHAPTER 3
OIL MARKET OUTLOOK H
I
G
H
L
I
G
H
T
S
Global primary demand for oil (excluding biofuels) in the Reference Scenario rises by 1% per year on average, from 85 million barrels per day in 2007 to 106 mb/d in 2030. This is a significant downward revision from last year’s Outlook, reflecting mainly the impact of much higher prices and slightly slower GDP growth. New government policies introduced in the past year also contribute to lower demand. All the increase in world oil demand comes from non-OECD countries. India sees the fastest growth, averaging 3.9% per year over the Outlook period, followed by China, at 3.5%. High as they are, these growth rates are still significantly lower than historic trends. Other emerging Asian economies and the Middle East also see rapid growth. By contrast, demand in all three OECD regions falls, due largely to declining non-transport demand. As a result of these trends, the share of OECD countries in global oil demand drops from 57% in 2007 to 43% in 2030. These oil-demand projections, combined with our oil-price assumptions, point to persistently high levels of spending on oil in both OECD and non-OECD countries. As a share of world GDP at market exchange rates, oil spending soared from a little over 1% in 1999 to around 4% in 2007, with serious implications for the economies of consuming countries. That share is projected to stabilise at around 5% over much of the Outlook period. For non-OECD countries, the share averages 6% to 7%. Most of the projected increase in world oil supply comes from OPEC countries, which hold the majority of the world’s remaining reserves of conventional oil. Their share of global output rises from 44% in 2007 to 51% in 2030. Their reserves are big enough for output to grow faster, but investment is assumed to be constrained, notably by conservative depletion policies and geopolitical factors.
The projected increase in global oil output hinges on adequate and timely investment. There remains a real risk of a supply crunch in the medium term as the gap between the capacity that is due to come on stream from current
Chapter 3 - Oil market outlook
91
© OECD/IEA, 2008
Although global oil production is not expected to peak before 2030, output of conventional crude oil and natural gas liquids levels off towards the end of the projection period. Non-conventional oil production, mainly from oil sands in Canada, continues to grow steadily, keeping total non-OPEC output broadly flat over the second half of the projection period.
projects and that needed to keep pace with demand widens sharply after 2010, squeezing spare capacity and driving up oil prices – possibly to new record highs. The volume of inter-regional oil trade increases by a third between 2007 and 2030. The Middle East, already the biggest exporting region, sees its net exports rise most. OECD Europe and Asia become even more dependent on imports over the projection period, but net imports drop in North America and in OECD Pacific. Though increased trade consolidates mutual dependence, it also enhances the risk of short-term supply interruptions, particularly as much of the additional oil imports will have to come from the Middle East and transit vulnerable maritime routes.
Demand Trends in primary oil demand Global primary oil demand1 is projected to grow by 1.0% per year on average in the Reference Scenario, from 85 million barrels per day in 2007 to 94 mb/d in 2015 and to 106 mb/d in 2030 (Table 3.1). The share of oil in global primary energy demand drops from 34% in 2006 to 30% in 2030. This is a significant, 10-mb/d downward revision from last year’s Outlook, reflecting the impact of much higher prices and slightly slower GDP growth. A number of new government policies introduced in the past year — notably moves in the United States and Europe to promote more fuel-efficient vehicles and encourage biofuels supply — also contribute to the reduction. The pace of demand growth slackens progressively over the projection period, from an average of 1.3% per year in 2007-2015 to 0.8% per year in 2015-2030.
1. Preliminary data on total oil demand by region only are available for 2007. The breakdown of oil demand by sector is available to 2006. Oil does not include biofuels derived from biomass, though transport demand for oil is modelled in a way that takes account of the use of biofuels. Regional totals do not include international marine bunkers. For these reasons, the oil projections in this report are not directly comparable with those published in the IEA’s Oil Market Report. See Annex B for a detailed definition of oil.
92
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Economic activity remains the principal driver of oil demand in all regions. Since 1980, each 1% per year increase in gross domestic product has been accompanied by a 0.3% rise in primary oil demand (Figure 3.1). This ratio — the oil intensity of GDP, or the amount of oil needed to produce one dollar of GDP — has fallen steadily over the last three decades. However, the decline has accelerated since 2004, mainly as a result of higher oil prices, which have encouraged conservation, more efficient oil use and switching to other fuels. In 2007, global oil intensity (expressed in purchasing power parities, or PPP) was barely half the level of the early 1970s.
Table 3.1
World primary oil demand* in the Reference Scenario (million barrels per day) 1980
2000
2007
2015
2030
2007-2030**
OECD
41.7
46.0
46.5
45.7
43.9
–0.2%
North America
20.9
23.3
24.6
23.9
23.9
–0.1%
United States
17.4
19.3
20.2
19.3
19.0
–0.3%
14.6
14.3
14.0
13.9
13.1
–0.3%
Europe Pacific
6.1
8.4
7.9
7.8
7.0
–0.5%
Japan
4.9
5.4
4.8
4.4
3.5
–1.4%
20.9
27.3
34.9
44.6
57.7
2.2%
Non-OECD E. Europe/Eurasia Russia Asia
9.5
4.4
4.8
5.7
5.9
0.9%
n.a.
2.7
2.8
3.3
3.4
0.7%
4.5
11.5
15.8
21.4
30.8
3.0%
China
2.0
4.7
7.5
11.3
16.6
3.5%
India
0.7
2.3
2.9
4.1
7.1
3.9%
Middle East
2.0
4.6
6.2
8.4
10.5
2.3%
Africa
1.3
2.3
2.9
3.2
3.7
1.0%
Latin America
3.5
4.5
5.2
5.9
6.8
1.2%
1.3
1.9
2.0
2.4
2.8
1.5%
2.3
3.0
3.8
4.1
4.7
1.0%
64.8
76.3
85.2
94.4
106.4
1.0%
n.a.
13.6
13.4
13.2
12.4
-0.3%
Brazil International marine bunkers World European Union
3
* Excludes biofuels demand, which, after adjusting for energy content, is projected to rise from just over 0.8 mb/d in 2007 to 2 mb/d in 2015 and 3.2 mb/d in 2030 in the Reference Scenario (see Chapter 7). ** Average annual rate of growth.
Figure 3.1
Change in world primary oil demand and real GDP growth
6%
GDP (PPP) Oil demand
4% 2% 0%
–4% –6% 1980
1985
Chapter 3 - Oil market outlook
1990
1995
2000
2005 2007
93
© OECD/IEA, 2008
–2%
The downward trend in global oil intensity is projected to continue through the projection period: by 2.2% per year on average between 2007 and 2030 compared with 2.1% in 1980-2007. At present, non-OECD countries are more oil-intensive than the OECD as a whole (using GDP on a PPP basis). By around 2015, their intensities more or less converge and then continue to decline at similar rates through to 2030 (Figure 3.2). The faster rate of decline in oil intensity in non-OECD countries in the first half of the projection period results from a relatively higher percentage increase in prices to end users there as subsidies are reduced on prices which carry a smaller burden of taxation (which, in OECD countries, cushions the impact on demand of changes in international prices). Each percentage-point increase in world GDP yields an increase in oil demand of 0.31% in 2015 and 0.24% in 2030.
toe of primary oil consumption per thousand dollars of GDP (2007 prices, PPP)
Figure 3.2
Oil intensity by region in the Reference Scenario
0.12
OECD Non-OECD
0.10 0.08 0.06 0.04 0.02 0 1980
1990
2000
2010
2020
2030
The apparent lack of responsiveness of global oil demand so far to the very substantial rise in international oil prices since the end of the 1990s and especially since 2004 contrasts sharply with the fall in demand that followed the two oil shocks of the 1970s (Figure 3.3). Five main factors explain this difference: The recent price surge has been driven in part by very strong economic growth worldwide, which has pushed up demand. In contrast, the oil shocks of the 1970s were largely supply-driven and, given the higher oil intensity of the world economy at the time, quickly depressed economic activity.
Price subsidies in many non-OECD countries, which prevent higher international crude oil and product prices from feeding through fully into final prices, thereby cushioning the impact on demand, have risen sharply (Box 3.1). The majority of oil consumers worldwide do not pay prices that fully reflect international market levels. By contrast, high excise-tax rates in most OECD countries mean that increases in international prices lead to much smaller proportionate increases in final prices. 94
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Opportunities for end users to switch from oil to other fuels are more limited than in the past, largely because of the bigger share of oil use that now goes to transport. Viable alternatives to gasoline and diesel remain limited for now.
The depreciation of the dollar against most major currencies has offset part of the increase in spot prices in recent years. Expressed in euros, for example, prices since the beginning of 2000 to the middle of 2008 rose only 57% as much as in dollars (Figure 3.4).
3
The increase in prices was very sudden in the first two oil shocks, whereas prices have risen in a much steadier manner since 2004.
Dollars per barrel (2007)
Figure 3.3
Global oil demand and the oil price in the three oil shocks
35
Increase in average IEA crude oil import price
30 25
Change in primary oil demand (right axis)
20 15 10
4%
5
2%
0
0% –2% –4% –6% 1973-1975
Index (January 2000 = 100)
Figure 3.4
1979-1981
–8%
2004-2006
Average IEA crude oil import price in different currencies
400
Dollar
350
Euro
300 250 200 150
50 0 2000
2001
Chapter 3 - Oil market outlook
2002
2003
2004
2005
2006
2007 2008
95
© OECD/IEA, 2008
100
Box 3.1
The impact on demand of removing oil subsidies
With oil-product prices subject to government price controls in the majority of non-OECD countries, the run-up in international prices in recent years has led to much higher subsidies in most of them because retail prices have been adjusted only with a lag and not to the full extent of the international price change. In many countries, these subsidies — which are typically either covered directly out of government budgets or are absorbed by national companies — are reaching unsustainable levels. For example, if prices had remained at mid-year levels Venezuela’s subsidy to oil products would have risen to around $14 billion in 2008, and the subsidy in Mexico, the only OECD country that still controls oil prices, to $20 billion — equal to several percentage points of GDP. Several countries are starting to address this problem. In early 2008, Chinese Taipei and Sri Lanka raised gasoline and diesel prices. In May, Indonesia raised the prices of kerosene, diesel and gasoline by between a quarter and a third, though these fuels remain heavily subsidised. In June, China announced a rise of around 18% in the pump prices of gasoline and diesel, while India raised pump prices and the retail prices of bottled liquefied petroleum gas (LPG) by between 8% and 17%. Malaysia raised gasoline prices by around 40% and diesel prices by two-thirds. Some other Asian countries, including Bangladesh and Vietnam, are expected to follow suit. Public resistance generally remains a major constraint on effective political action to reduce and remove subsidies in non-OECD countries. Indeed, some countries with subsidies were considering lowering controlled prices in response to the drop in international oil prices in August and early September 2008.
There are, nonetheless, signs that high oil prices are beginning to affect oil demand, directly and indirectly, through their depressive impact on economic activity. The slowdown in demand has been most apparent in OECD countries, where prices have generally risen the most. Preliminary data point to a more than 1% drop in OECD inland deliveries (oil products supplied by refineries, pipelines and terminals) in June 2008 compared with June 2007. Consumption of transport fuels has fallen sharply in some countries, including the United States, as people drive less and begin to switch to smaller cars. For example, new car sales in the United Kingdom in August 2008 fell to their lowest level in 42 years. Sales of large sports utility vehicles have fallen in most 96
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Recent evidence and our estimates of the price elasticities of demand for oil products imply that moves to cut subsidies would have a significant effect on overall oil demand, especially in the long term. The demand response would vary markedly by country, according to the extent of the subsidy and the oil products concerned. A 90% jump in the pump prices of gasoline and diesel in Indonesia in October 2005, for example, caused consumption to drop by 13% in 2006. The recent cut in Malaysian subsidies is expected to lead to a drop in road-fuel use of around 5% in the next year or so. These moves will help to alleviate the upward pressure on international oil prices and domestic spending on transport in the coming years, as higher prices influence decisions to purchase cars and oil-consuming equipment.
countries, especially in the United States. Non-OECD oil demand, in contrast, continues to grow steadily, largely because incomes have continued to grow and prices in most countries have not risen so fast thanks to subsidies. As a result of these factors, oil demand, in the short term at least, is relatively insensitive to movements in international crude oil prices. The weighted average crude oil price elasticity of total oil demand across all regions is estimated at -0.03 in the short term (in the first year following the price change) and -0.15 in the long term, based on econometric analysis of historical demand trends (IEA, 2006).2 Using these estimates, we calculate that, had international prices not risen since 2003, global primary oil demand would have grown on average by 1.9% per year in the four years to 2007 — a mere 0.1 percentage points more than it actually did — on the assumption that nothing else was different. On this basis, world demand would, therefore, have been about 300 kb/d higher in 2007.
3
Regional trends All the increase in world oil demand between 2007 and 2030 comes from nonOECD countries in the Reference Scenario. Their demand rises by 22.8 mb/d, offsetting a 2.5 mb/d fall in the OECD (international bunker demand rises by almost 1 mb/d). India sees the fastest rate of growth, averaging 3.9% per year over the Outlook period, followed by China, at 3.5%. High as they are, these growth rates are still significantly lower than historic trends. India’s oil use grew by 5.6% per year between 1980 and 2007 (quadrupling), while Chinese oil use increased almost as fast (from a larger base). In volumetric terms, China remains the single biggest contributor to the growth in world oil demand, accounting for 43% of the total projected increase in 2007-2030 (Figure 3.5). Other emerging Asian Figure 3.5
Change in primary oil demand by region in the Reference Scenario, 2007-2030
OECD Pacific OECD Europe OECD North America Africa E. Europe/Eurasia Latin America Other Asia India Middle East China 0
2
4
6
8
10 mb/d
2. The price elasticity of demand for oil products based on final prices is significantly higher and more homogeneous, as the impact of differences in tax and subsidy policies is stripped out. The final price elasticity of road-transport fuels, for example, is estimated at -0.15 in the short run (after one year) and -0.44 in the long run (IEA, 2006).
Chapter 3 - Oil market outlook
97
© OECD/IEA, 2008
–2
economies and the Middle East also see rapid rates of growth. The latter region has emerged as a major oil consuming as well as producing region, on the back of a booming economy (helped by high oil prices) and heavily subsidised prices. Middle East countries account for 20% of the growth in oil demand over the projection period. Demand in all three OECD regions, by contrast, falls, most heavily in volume terms in the Pacific region and Europe. As a result of these trends, the non-OECD countries’ share of global oil demand (excluding international marine bunkers) rises from 43% in 2007 to 57% in 2030. The differences in per-capita oil consumption across regions remain striking. The population of the Middle East reaches only 281 million in 2030, compared with nearly 1.5 billion in China. Yet China’s oil use is only half as high as the Middle East’s. Percapita oil consumption reaches 13.6 barrels in the Middle East in 2030, compared with only 4.2 barrels in China and 12.3 barrels in the OECD.
Sectoral trends Around three-quarters of the projected increase in oil demand worldwide comes from the transport sector, the sector least responsive, in the short term, to price changes (Figure 3.6). In the OECD, oil use falls in all sectors except transport (where it is essentially flat).
Mtoe
Figure 3.6
Incremental oil demand by sector in the Reference Scenario, 2006-2030
800
Rest of world Other Asia
600
India China
400
OECD 200 0 –200 Transport
Industry
Non-energy use
Other*
Residential and industrial demand falls most heavily. In non-OECD countries, transport is the biggest contributor to oil-demand growth, but the industrial, non-energy use, and residential, services and agriculture sectors also account for significant shares. The transport sector accounts for 57% of global primary oil consumption in 2030, compared 98
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
* Includes residential, services, agriculture, power generation and other energy sectors.
with 52% now and 38% in 1980 (Table 3.2). Oil-based fuels continue to dominate transport energy demand, though their overall share falls slightly, from 94% in 2006 to 92% in 2030, due to growing use of biofuels (see Chapter 7). Table 3.2
3
Share of transport in primary oil demand by region in the Reference Scenario (%) 1980
2000
2006
2015
2030
OECD
40
54
57
60
62
North America
51
63
64
68
70
United States
52
66
66
71
72
Europe
31
50
53
57
58
Pacific
27
38
41
44
44
Japan
23
36
37
39
38
Non-OECD
29
39
41
44
50
E. Europe/Eurasia
24
37
42
45
45
n.a.
36
40
44
44
25
36
37
42
51
China
17
32
36
43
54
India
37
29
27
32
45
32
36
40
39
43
Russia Asia
Middle East Africa
46
51
50
51
54
Latin America
39
47
48
52
54
44
47
52
55
56
38
51
52
54
57
n.a.
51
54
58
59
Brazil World European Union
3. There are signs that consumers are beginning to switch to smaller, more fuel-efficient cars in response to higher oil prices, especially in countries with lower tax rates and where prices have risen most in percentage terms. In the United States, for example, which has the lowest rates of tax on gasoline in the OECD, sales of sports utility vehicles have fallen heavily in the last two to three years.
Chapter 3 - Oil market outlook
99
© OECD/IEA, 2008
Despite continuing improvements in average vehicle fuel efficiency, spurred in part by high oil prices,3 the sheer growth of the vehicle fleet — especially in non-OECD countries — is expected to continue to push up total oil use for transport purposes. There is not expected to be any major shift away from conventionally fuelled vehicles before 2030, though the penetration of hybrid-electric cars, including plug-ins (see the Spotlight below), is expected to rise, contributing to an overall improvement in fuel economy. Globally, passenger-car ownership is assumed to grow at 3.6% through the projection period, outpacing GDP slightly. The total light-duty vehicle stock rises from an estimated 650 million in 2005 to about 1.4 billion by 2030. The biggest increase in absolute terms is expected to occur in China, which accounts for almost one-third of
the global increase in cars (Figure 3.7). Most of the rest of the increase comes from other non-OECD countries. The average size and power of new cars in these countries is expected to be significantly lower than in the OECD at present, which accounts for part of the global improvement in fuel economy.
Figure 3.7
Light-duty vehicle stock by region in the Reference Scenario
Africa
2005
Middle East
2030
Other Asia OECD Pacific Latin America E. Europe/Eurasia India OECD North America China OECD Europe 0
50
100
150
200
250
300 Million
Power generation currently accounts for around 7% of global primary oil consumption, down from 14% in 1980. Its share is projected to decline further, to around 4% in 2030. The volume burned in power stations also drops, from 277 Mtoe to about 204 Mtoe. By the end of the projection period, close to 85% of oil use for power is in non-OECD regions — about 40% of which is in the Middle East. Investment in refinery upgrading units is expected to continue to reduce the proportion of residual heavy fuel oil — which makes up the bulk of the oil used by power stations and by ships — in the output of refined products. The fuel oil that is produced is increasingly concentrated in the power sector. While the share of oil in overall fuel inputs to power generation is projected to fall sharply over the projection period, there may be pronounced short-term swings in response to fluctuations in relative oil, gas and coal prices, delays in bringing generating capacity on line, and unplanned outages of other base-load plants. A shortfall in coal-fired capacity in China in 2004 led to a surge in oil burning, which contributed significantly to a jump in world oil demand and to upward pressures on international prices.
These oil-demand projections, combined with our oil-price assumptions (described in Chapter 1), point to persistently high levels of spending on oil in both OECD and non-OECD countries. As a share of world GDP at market exchange rates, oil spending 100
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Implications for spending on oil
S P O T L I G H T
What is the scope for switching from oil-fuelled to electric vehicles?
Source: IEA (2008).
Chapter 3 - Oil market outlook
101
3
© OECD/IEA, 2008
Electric vehicles continue to struggle to compete with conventional oil-fuelled cars and trucks. Although fuel costs for owners of electric vehicles are lower, the savings are not big enough to offset the much higher prices of the vehicles themselves. Fuel subsidies in many non-OECD countries also undermine the attractiveness of electric vehicles. Yet electric-vehicle technology is advancing rapidly. Vehicle hybridisation, involving the addition of an electric motor and an energy-storage system (typically a battery) to a conventional engine/fuel system, has attracted most investment and has already proved commercially successful — in spite of relatively high costs. Further improvements to storage systems are necessary to boost efficiency and lower costs: despite significant progress in recent years, even the best lithium-ion batteries available today suffer from inadequate performance and high costs. Ultra-capacitors, which store energy in charged electrodes rather than in an electrolyte, are increasingly being seen as a complement to batteries; they store less energy per unit weight than batteries but are able to deliver energy more quickly. Research into these technologies is expected to yield further major improvements in the coming years. In the longer term, plug-in hybrids, fully electric vehicles and hydrogen fuel cells appear to offer the greatest potential for reducing or eliminating the need for oilbased fuels. Plug-in hybrids, in which the electric battery is recharged off the grid as well as through the vehicle’s internal recharging system, combine the efficiency advantages of using an electric motor for short distances with the convenience and longer range of an internal combustion engine. How quickly plug-ins become commercially viable will hinge on substantial improvements in the performance of electric batteries. The prospects for fully electric vehicles are less certain, as battery capacity needs to be even greater than for hybrids. The need for quick recharging and related infrastructure is an additional barrier. A lot of research effort continues to be focused on fuel cells powered with hydrogen (stored on-board or supplied from a reformer using a hydrogen-rich feedstock). But, as with electric vehicles, a number of technical hurdles remain to be overcome in order to lower costs and make fuel-cell vehicles commercially viable. Neither fully electric nor hydrogen-powered fuel-cell vehicles are expected to become widely available before 2030 unless research efforts are stepped up and technological breakthroughs are achieved (limited penetration of the former is assumed only in the 450 Policy Scenario). They could, however, be commercialised widely before 2050. In the IEA’s most recent Energy Technology Perspectives, fuel-cell vehicles account for 33% to 50% of total car sales in the OECD and half those in the rest of the world by 2050 in scenarios that assume a significant increase in research spending over the next decade and rapid reductions in unit costs.
soared from a little over 1% in 1998 to around 4% in 2007. On current trends, that share is set to approach 6% in 2008, stabilising at around 5% over much of the Outlook period (Figure 3.8). For non-OECD countries, the share is even higher, reaching 8% in 2008 and falling back to above 6% by 2030. Only in the early 1980s has this share been so high in both OECD and non-OECD regions. Figure 3.8
Share of oil spending in real GDP at market exchange rates in the Reference Scenario
12%
World OECD
10%
Non-OECD 8% 6% 4% 2% 0 1971
1980
1990
2000
2010
2020
2030
Production4 World oil supply is projected to rise from 84 mb/d in 2007 to 106 mb/d in 2030. Netting out processing gains in refining, production reaches 104 mb/d (Table 3.3). The bulk of the projected increase in world oil supply is projected to come from OPEC countries, where most of the world’s remaining proven reserves of conventional crude oil are to be found. Their share of global output rises from 44% in 2007 to 51% in 2030 — though this is still below the historical peak of 53% in 1973. Their reserves are, in principle, big enough and development costs low enough for output to grow faster than this. However, investment is assumed to be constrained by several factors, including conservative depletion policies and geopolitics (see Chapter 13).
4. This section summarises the Reference Scenario projections of global oil production, which are described in detail in Chapter 11 in Part B.
102
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Non-OPEC conventional oil production is projected to peak by around the middle of the next decade and then to decline slowly through to the end of the projection period. This is largely offset by rising non-conventional output, which keeps total non-OPEC output broadly flat over the second half of the projection period. Higher prices have stimulated new upstream investment in the last few years, though much of the rise in this upstream expenditure is due to higher costs (see Chapter 13). This investment will boost capacity
in the near term, more than offsetting natural production declines at existing fields (see Chapter 10). But dwindling new discoveries and a fall in the size of new fields are expected to drive up marginal development costs. As a result, oil prices are assumed to rise to stimulate the necessary investment. Production has already peaked in most nonOPEC countries and will peak in most others before 2030. Russian output is projected to edge up by 2015 but then to drift lower towards the end of the Outlook period. Table 3.3
3
World oil supply in the Reference Scenario (million barrels per day)
Non-OPEC OECD North America
1980
2000
2007
2015
2030
2007-2030*
35.5
42.9
46.3
47.6
50.9
0.4%
17.3
21.8
19.3
18.6
20.8
0.3%
14.2
14.1
13.8
14.6
17.9
1.1% -3.5%
Europe
2.6
6.8
4.9
3.4
2.2
Pacific
0.5
0.9
0.6
0.6
0.7
0.6%
12.1
8.1
12.9
14.3
16.6
1.1%
10.8
6.5
10.1
10.4
9.7
-0.2% -0.4%
E. Europe/Eurasia Russia Asia
2.9
5.6
6.4
6.1
5.8
China
2.1
3.2
3.7
3.6
4.1
0.5%
India
0.2
0.7
0.8
0.8
0.5
-1.8%
Middle East
0.6
2.1
1.7
1.4
1.1
-1.8%
Africa
1.1
2.1
2.5
2.1
1.9
-1.1%
Latin America
1.6
3.2
3.5
5.1
4.6
1.2%
OPEC**
28.1
32.1
35.9
44.4
52.9
1.7%
Middle East
19.2
21.3
23.7
30.7
37.9
2.1%
1.7
1.7
2.1
2.3
2.6
1.0%
65.2
76.8
84.3
94.4
106.4
1.0%
63.1
73.8
80.7
87.4
95.0
0.7%
0.4
1.2
1.6
4.6
8.8
7.7%
Processing gains World Conventional*** Non-conventional****
Although global oil production is not expected to peak before 2030, conventional crude oil production (including natural gas liquids and enhanced oil recovery) is projected to level off towards the end of the projection period. A growing share of the increase in world output comes from non-conventional sources, mainly Canadian oil sands, extra-heavy oil, gas-to-liquids and coal-to-liquids (Figure 3.9). These increases in global oil output hinge on adequate and timely investment, which remains a critical uncertainty (see Chapter 13). There remains a real risk of an oil-supply crunch in the medium term resulting from under-investment (Box 3.2). Chapter 3 - Oil market outlook
103
© OECD/IEA, 2008
* Average annual rate of growth. ** Includes Angola, which joined OPEC at the beginning of 2007. *** Includes conventional crude oil, condensates, natural gas liquids (NGLs) and extra-heavy oil from Venezuela. ****Extra-heavy oil (excluding Venezuela), oil sands, chemical additives, gas-to-liquids and coal-to-liquids. Biofuels are not included.
Figure 3.9
World oil production by source in the Reference Scenario
mb/d
120
Non-conventional
100 80
8% 2%
60 56%
42%
Middle East OPEC
54%
40 20
Other OPEC
5%
2% 47%
Non-OPEC
15% 14%
15%
28%
29%
2000
2007
14%
33%
36%
2015
2030
0 Note: Excludes processing gains. Conventional oil includes crude oil, condensates, natural gas liquids (NGLs) and extra-heavy oil from Venezuela.
Box 3.2
Averting a post-2010 oil-supply crunch
The sheer scale of the investment needed raises questions about whether all of the additional capacity we project will be needed will actually occur. If actual capacity additions fall short of this amount, spare production capacity would be squeezed and oil prices would undoubtedly rise — possibly to new record highs (Jesse and van der Linde, 2008; Stevens, 2008). These concerns are behind the initiative of the UK prime minister to invite major oil producing and consuming countries to a meeting in London in December 2008 to discuss ways of lowering barriers to investment, following up a meeting in Jeddah (in which the IEA participated) called by Saudi Arabia in June. Boosting investment in the resource-rich countries, which, in many cases, is restricted, will be of critical importance to averting a supply crunch.
104
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Despite shortages of equipment and labour, and lengthening delays in bringing new projects to completion, the current wave of upstream investment looks set to provide significant boost to oil-production capacity in the next two to three years. Based on our analysis of upstream investment, we expect gross capacity additions worldwide to cover demand growth and output declines from existing fields to 2010, pushing up spare capacity modestly (see Chapter 13 for details). However, capacity additions from current projects tail off after 2010. This largely reflects the upstream development cycle: many new projects will undoubtedly be sanctioned in the next two or three years as oil companies complete existing projects and move on to new ones. But the gap between what is currently being built and what will be needed to keep pace with demand is nonetheless set to widen sharply after 2010 (see Chapter 11). Around 7 mb/d of additional capacity needs to be brought on stream by 2015 (over and above the capacity that is already in the pipeline from current projects), most of which will need to be sanctioned within the next two years or so.
Inter-regional trade The mismatch between demand growth and the sources of incremental production means that inter-regional trade in oil (crude oil, NGLs and refined products) increases sharply during the projection period in the Reference Scenario. The volume of trade reaches 55 mb/d in 2030 — over half of global oil production and over 35% more than at present (Table 3.4). Net exports from the Middle East, already the biggest exporting region, rise from 19.9 mb/d in 2007 to 28.5 mb/d in 2030, representing 52% of global trade in 2030, up from 49% today. Net exports from Africa and Eastern Europe/Eurasia (thanks to the Caspian region) also continue to expand steadily. Brazil contributes most to the increase in net exports from Latin America in the early part of the Outlook period, and Venezuela thereafter.
3
Of the net oil-importing regions today — the three OECD regions and non-OECD Asia — Europe and Asia become even more dependent on imports over the projection period (Figure 3.10). By contrast, net imports into North America drop from 10.7 mb/d to 5.9 mb/d, largely thanks to rising Canadian oil sands production and a sharp slowdown in demand. Net imports into the United States alone drop much less, from 13.2 mb/d in 2007 to 11.9 mb/d in 2030 — equal to 63% of total consumption. Import dependency in the OECD Pacific region also falls slightly, to 90% in 2030 compared with 92% currently. OECD Europe, by contrast, sees a sharp rise in net import dependence, from 65% to 84%, as a result of the projected rapid decline in North Sea production. Net imports into the OECD as a whole decline as a share of total oil demand from 58% in 2007 to 52% in 2030. In Asia, the increase in dependence is dramatic in India, where imports meet almost all the country’s needs in 2030, and in China, which became a net oil importer only in 1993. China’s imports rise from 3.8 mb/d in 2007 to 12.5 mb/d in 2030, equal to 75% of China’s total oil demand, compared with 50% in 2007.
Oil-import dependence by major importing region in the Reference Scenario
OECD North America
2007
10.7 mb/d
5.9 mb/d
13.2 mb/d 11.9 mb/d
United States
2030
3.8 mb/d
China
12.5 mb/d 8.4 mb/d
Other Asia
24.7 mb/d 9.1 mb/d
OECD Europe
11.0 mb/d 7.3 mb/d 6.2 mb/d 11.0 mb/d 11.5 mb/d
OECD Pacific European Union 2.1 mb/d
India
6.6 mb/d
20%
30%
Chapter 3 - Oil market outlook
40%
50%
60%
70%
80%
90%
100%
105
© OECD/IEA, 2008
Figure 3.10
Increased trade consolidates global interdependence, but the risk to consuming countries of short-term supply interruptions grows as geographic supply diversity is actually reduced, increasing reliance on a few supply routes. Much of the additional oil imports will come from the Middle East, the scene of most of the biggest supply disruptions in the past, and will transit vulnerable maritime routes to both eastern and western markets. Almost all Middle East oil and gas exports, for example, transit the Straits of Hormuz at the mouth of the Arabian Gulf (Figure 3.11). Any supply disruption would drive up prices to all consuming countries, regardless of where they obtain their oil. The growing participation of China, India and other emerging countries in international oil trade amplifies the importance of their contribution to collective efforts to enhance global energy security (IEA, 2007).
Table 3.4
Net inter-regional oil trade in the Reference Scenario (million barrels per day) 1980
2000
2007
2015
2030 -23.1
Net importers OECD
-24.4
-24.2
-27.1
-27.1
North America
-6.7
-9.2
-10.7
-9.3
-5.9
United States
-7.2
-11.3
-13.2
-12.3
-11.9 -11.0
-12.0
-7.5
-9.1
-10.6
Pacific
Europe
-5.6
-7.5
-7.3
-7.2
-6.2
Japan
-4.9
-5.4
-4.8
-4.4
-3.5
Non-OECD Asia
0.0
-4.6
-8.4
-14.5
-24.7
China
0.2
-1.4
-3.8
-7.7
-12.5
India
-0.5
-1.6
-2.1
-3.3
-6.6
17.7
18.9
19.9
23.9
28.5
4.9
5.6
7.5
8.7
9.1 1.7
Net exporters Middle East Africa Latin America Brazil E. Europe/Eurasia Russia Total trade European Union (imports)
0.3
2.3
1.5
2.3
-1.1
-0.7
-0.2
1.2
0.6
2.6
4.0
8.1
8.6
10.7
n.a.
3.8
7.2
7.0
6.3
27.6
34.7
40.7
46.6
55.1
n.a.
-10.1
-11.0
-11.8
-11.5
106
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Note: Trade between WEO regions only. Positive figures denote exports; negative figures imports.
Chapter 3 - Oil market outlook
5.5
6.0
4.9
4.2
5.3
Oil flow, 2006 (mb/d)
Bab el-Mandab
4.6
Suez
5.1
from West Africa
to Australia & New Zealand
Hormuz
20.7 20.5 23.0
107
© OECD/IEA, 2008
The boundaries and names shown and the designations used on maps included in this publication do not imply official endorsement or acceptance by the IEA.
to Atlantic basin markets
3.8
17.0
Oil export flows from the Middle East
Share of world oil demand (%) 2006 2015 (Reference Scenario) 2030 (Reference Scenario)
3.9
Figure 3.11
to Far East
Malacca
12.0
14.3 15.3 16.7
3
© OECD/IEA, 2008
CHAPTER 4
NATURAL GAS MARKET OUTLOOK I
G
H
L
I
G
H
T
S
World primary demand for natural gas expands by just over half between 2006 and 2030 in the Reference Scenario to 4.4 trillion cubic metres, a rate of increase of 1.8% per year. The share of gas in total world primary energy demand increases marginally, from 21% in 2006 to 22% in 2030. The projected growth in gas use is nonetheless slower than in last year’s Outlook — the result of higher assumed prices and less-rapid growth in gas-fired power generation. The bulk of the increase in global gas use over the projection period — more than three-quarters in total — comes from non-OECD regions, especially those that are well endowed with gas resources. Gas demand is projected to grow most in absolute terms in the Middle East. The pace of demand growth is fastest in China. Despite their much less rapid economic growth, North America and Europe still contribute a fifth of the global increase in gas demand. The power sector accounts for 57% of the projected increase in world gas demand to 2030, pushing up its share of global gas use from 39% now to 45%. The power sector is the main driver of gas demand in almost all regions, especially in non-OECD countries where electricity demand is projected to rise most rapidly. Despite the recent surge in price, gas remains competitive with other fuels in many cases, especially for midload generation and where coal has to be imported. Gas resources are sufficient to meet the projected increase in global demand, but production is set to become much more concentrated in the most resource-rich regions. Some 46% of the projected growth in world gas production in 2006-2030 comes from the Middle East, its output tripling to around 1 tcm by 2030. About 60% of the region’s incremental output is consumed locally, mainly in power stations. Most of the remaining increase in world output is provided by Africa and Eastern Europe and Eurasia (mainly Russia). These capacity expansions hinge on timely investment. Inter-regional natural gas trade is projected to more than double over the projection period, from 441 billion cubic metres in 2006 to just over 1 tcm in 2030. Imports rise in all the regions except non-Russia Eurasia that are currently net importers of gas, both in volume and as a share of their total gas consumption. The European Union sees the biggest increase in import volumes. Most of the growth in gas exports over 2006-2030 comes from the Middle East and Africa. Together, they account for about 60% of total exports in 2030. Russia and the Caspian/Central Asian countries combined remain the other main exporting region. Most of the increase in inter-regional trade is in the form of liquefied natural gas, its share of trade rising from 52% in 2006 to 69% in 2030. Liquefaction capacity is set to expand markedly through to early in the 2010s, but a shortage could emerge thereafter if a wave of new investment is not sanctioned soon.
Chapter 4 - Natural gas market outlook
109
© OECD/IEA, 2008
H
Demand Global demand for natural gas is projected to increase by 52% between 2006 and 2030, from 2 916 bcm to 4 434 bcm (Table 4.1). Demand grows at an average annual rate of 1.8%, 0.3 percentage points lower than was projected in last year’s Outlook. The share of gas in total world primary energy demand increases marginally, from 21% in 2006 to 22% in 2030. In 1980, the share of gas was only 17%. Table 4.1
World primary demand for natural gas in the Reference Scenario (billion cubic metres) 1980
2000
2006
2015
2030
2006-2030*
OECD
958
1 407
1 465
1 645
1 827
0.9%
North America
659
799
766
848
908
0.7%
United States
581
669
611
652
631
0.1%
Europe
264
478
541
614
694
1.0%
Pacific
35
130
158
183
225
1.5%
Japan
25
82
94
104
128
1.3%
Non-OECD
559
1 135
1 451
1 867
2 607
2.5%
E. Europe/Eurasia
438
606
676
779
846
0.9%
Russia
n.a.
395
444
507
524
0.7%
36
185
285
414
666
3.6%
China
14
28
58
121
221
5.8%
India
1
25
38
57
117
4.8%
36
182
276
378
676
3.8%
Asia
Middle East Africa
14
62
90
124
168
2.6%
Latin America
36
100
124
174
252
3.0%
Brazil World European Union
1
9
21
32
46
3.3%
1 517
2 541
2 916
3 512
4 434
1.8%
n.a.
482
532
606
681
1.0%
The projected rate of growth in global gas demand is lower than over the past quarter of a century: demand rose by 2.6% per year between 1980 and 2006. Demand began to slow at the turn of the century, rising by only 2.3% per year over 2000-2006. Warmer winter weather across the northern hemisphere, coupled with higher prices, largely offset the impact of relatively rapid economic growth worldwide over that period. The increase in prices in recent years has affected gas demand more than has been the case for oil, mainly because there is more scope for consumers to switch at short notice to other fuels, especially in power generation and industry (see Spotlight below). The much higher level of gas prices, in absolute terms and relative to coal prices (see Chapter 1), is the main reason for the downward revision in projected gas-demand growth. Slower GDP growth, especially in the big gas-consuming OECD regions, also contributes. 110
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
* Average annual rate of growth.
S P O T L I G H T
Are high prices choking off demand for gas? The scope for gas consumers to reduce their consumption of gas in the face of high prices varies by region and by sector. Most residential consumers have little option but to pay the higher cost, as they rarely have any back up for space and water heating, and for cooking. Moreover, switching to an alternative fuel when installing new equipment may not be attractive since heating oil and electricity prices have also risen in most cases. But some gas-fired power plants and industrial boilers can be switched to other fuels at short notice (environmental regulations permitting), usually heavy fuel oil in the case of conventional steam boilers and distillate in the case of gas turbines. In addition, some utilities and manufacturers maintain back-up capacity that allows them to switch quickly to a cheaper fuel. Good statistics on the amount of such flexible capacity and the extent of switching in practice are rare outside the United States. The most recent survey of switching capacity in US manufacturing industry found that almost one-fifth of actual gas consumption could be avoided by switching to other fuels — a slightly lower share than in previous surveys (DOE/EIA, 2002). Roughly one-third of US power plants that use gas as the primary fuel were also able to run on oil products in 2007, even though most of the new gas-fired capacity added at the beginning of the current decade cannot use oil as a back up or alternative fuel. There is evidence that the surge in gas prices in the United States and other OECD countries has prompted some switching from gas to oil and other fuels in recent years. Usually, this switching has been temporary. The share of gas in power generation in the United Kingdom, for example, dropped by three percentage points to 36% in 2006 in response to a spike in gas prices relative to coal prices, but rebounded to 42% in 2007 with higher coal prices. In some cases, where no fuelswitching capacity exists, industrial production has stopped altogether — sometimes permanently. Since 2000, 26 of the 56 nitrogen fertilizer plants in the United States have shut down as a result of the higher cost of gas. Natural gas accounts for up to 90% of the production cost of ammonia — the raw material for all nitrogen fertilizers. The use of gas as a petrochemical feedstock is growing much faster in non-OECD countries, especially the Middle East, with low-cost local supplies. The decline in the gas intensity of economic activity worldwide has gathered pace since the beginning of the current decade, suggesting that higher prices are indeed tempering the growth in demand for gas (Figure 4.1). The decline in gas intensity has accelerated fastest on average in non-OECD countries — even though, in many cases, price controls have prevented much of the increase in market prices from feeding through into retail prices. Gas is subsidised in most non-OECD countries — in some cases, heavily. In Russia, for example, gas subsidies to households and industry totalled close to $30 billion in 2007 (see Chapter 1). The reduction of these subsidies contributes to the acceleration of the decline in gas intensity in non-OECD countries over the period 2006-2030.
Chapter 4 - Natural gas market outlook
111
© OECD/IEA, 2008
Sources: IEA databases; Department of Energy/Energy Information Administration (www.eia.doe. gov/cneaf/electricity/epa/epa_sum.html); US Fertilizer Institute (www.tfi.org).
4
toe of primary gas consumption per thousand dollars of GDP (2007, PPP)
Figure 4.1
Change in gas intensity by region in the Reference Scenario
0.010
OECD
0.005
Non-OECD World
0 –0.005 –0.010 –0.015 –0.020 –0.025 1980-2000
2000-2006
2006-2030
Regional trends As with most other primary energy sources, the bulk of the increase in global gas use over the projection period — 76% in total — comes from non-OECD regions, especially those that are well endowed with gas resources (Figure 4.2). Gas demand is projected to grow most in absolute terms in the Middle East, where it jumps from 276 bcm in 2006 to 676 bcm in 2030 — an increase of 145%. Rising local availability of gas supplies is expected to fuel this expansion in demand, most of which comes from power generation and petrochemicals (for use as feedstock and heat generation). By 2030, the Middle East accounts for 15% of world gas consumption, compared with only 9% today and a mere 2% in 1980. Demand also grows strongly in non-OECD Asia, where gas captures market share from coal in industry. The pace of demand growth is fastest in
Figure 4.2
Increase in primary demand for natural gas by region in the Reference Scenario 1980-2006
Middle East
2006-2030
Asia E. Europe/Eurasia OECD Europe OECD North America Latin America
OECD Pacific 0
112
50
100
150
200
250
300
350 400 450 Billion cubic metres
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Africa
China, at close to 6% per year, and India, at nearly 5% per year. Yet the two countries’ combined share of world demand reaches only 8% in 2030, up from 3% in 2006, as they start from a low base. Demand rises much less quickly in Russia and in other Eastern European/Eurasian countries, mainly due to assumed progress in raising final prices to market levels, improvements in end-use energy efficiency, reduced waste and saturation effects.
4
Despite their much less rapid economic growth, North America and Europe both still account for a fifth of the overall growth in world gas use. With the exception of Eastern Europe/Eurasia and the Middle East, per-capita consumption in North America and Europe remains well above that of all other regions, for climate, economic and policy reasons. In many cases, gas continues to be favoured over coal and oil for environmental reasons, especially in power generation, where gas can be used in highly efficient combined-cycle gas turbines (see below). In Europe, carbon penalties under the EU Emissions Trading Scheme help gas to compete against more carbon-intensive coal in the power sector and heavy industry.
Sectoral trends Gas use is projected to become increasingly concentrated in power generation. The power sector accounts for 57% of the increase in world gas demand over the projection period (Figure 4.3).1 As a result, the power sector’s share of global gas use rises from 39% in 2006 to 45% in 2030. The power sector is the main driver of gas demand in almost all regions, especially in non-OECD countries where electricity demand is projected to rise most rapidly. Despite the recent surge in price, in many cases gas remains competitive with other fuels, especially for mid-load generation and where coal has to
Figure 4.3
Increase in world primary natural gas demand by sector in the Reference Scenario, 2006-2030
Gas to liquids
% = average annual rate of growth
12.3%
Other sectors
1.0%
Industry
1.4%
Residential, services and agriculture
1.2% 2.4%
Power generation 0
200
400
600
800
1 000
1. The projected growth in global gas use in power generation, while strong, is slower than in last year’s Outlook, mainly because of less-rapid growth in demand for electricity and an increase in the cost of gasfired generation relative to other power-generation technologies.
Chapter 4 - Natural gas market outlook
113
© OECD/IEA, 2008
Billion cubic metres
be imported (see Chapter 6). The capital costs of combined-cycle gas turbines, which are much more efficient than coal-based power technologies, are lower than for coalfired power plants and the construction lead-times shorter. These factors compensate for higher fuel costs. In Europe, carbon penalties give gas an added cost advantage over coal for base-load generation. Nonetheless, renewables-based power technologies are expected to become increasingly competitive against gas over the projection period (see Chapter 7), slowing the rate of construction of gas-fired plants. Almost all of the rest of the increase in world primary gas demand to 2030 comes from end-use sectors, notably industry and the residential, services and agricultural sectors. Industry remains the single largest gas-consuming end-use sector, with demand projected to grow at 1.4% per year over 2006-2030. Industry’s share of total final gas demand rises from 35% in 2006 to 36% in 2030. Nonetheless, the share of gas in total industrial energy consumption falls slightly — mainly due to the increasing penetration of electricity, especially in non-OECD countries. Industrial gas use in the OECD barely increases, due to sluggish growth in heavy industrial production. As in industry, demand for gas in the residential, services and agricultural sectors grows relatively slowly. In OECD countries, the per-capita use of gas for space and water heating and for cooking is approaching saturation levels. With population growing only very slowly, this holds back the overall increase in gas demand in the residential, services and agricultural sectors to only 0.7% per year over the projection period. Gas consumption in these sectors grows more quickly, by an average of 2.2% per year in the non-OECD countries as a whole. Demand growth is particularly rapid in Asia. Gas-to-liquids (GTL) plants, which convert natural gas into high-quality oil products (as well as lubricants and petrochemical products), account for a small but growing share of global natural gas demand (Figure 4.4). The consumption of gas in such plants, mainly located in the Middle East and Africa, rises from only 5 bcm in 2006 to 33 bcm by 2015 and over 50 bcm in 2030. There are three commercial-scale GTL plants operating today: Oryx in Qatar, which started up in 2006; Shell’s Bintulu plant in Malaysia which came on stream in 1993; and the Mossgas plant in South Africa, which has been World primary natural gas demand by sector in the Reference Scenario
5 000
Other sectors Gas to liquids
4 000
Industry
52% growth 3 000
Residential, services, and agriculture
2 000
Power generation
1 000 0 1990
114
2000
2006
2015
2030
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Billion cubic metres
Figure 4.4
operating since 1990. Two more plants are under construction and others are planned. The long-term prospects for GTL hinge critically on bringing down costs, which have risen sharply in recent years, and on raising the thermal efficiency of the conversion process (see Chapter 11).
Production2 4
The proximity of reserves to markets and the level of production costs will continue to drive regional prospects for gas supply. Gas transportation, whether by pipeline or as liquefied natural gas (LNG), remains very expensive and usually represents a significant share of the overall cost of gas delivered to consumers. A substantial part of the world’s gas resources is located far from the main centres of demand and cannot yet be extracted profitably. Those resources are unquestionably large enough to meet the projected increase in global demand in the Reference Scenario. It is assumed here that the necessary investment in the infrastructure required to bring them to market will be made in a timely manner, though this is an area of risk (Box 4.1). Some 46% of the projected growth in world gas production over the Outlook period comes from the Middle East, its output tripling to around 1 tcm by 2030 (Table 4.2 and Figure 4.5). That region holds the largest reserves and has the lowest World natural gas production in the Reference Scenario (billion cubic metres) 1980
2000
2006
2015
2030
2006-2030*
OECD
889
1 107
1 117
1 149
1 086
-0.1%
North America
657
763
761
795
765
0.0%
United States
554
544
524
535
515
-0.1%
Europe
219
302
305
282
217
-1.4%
Pacific
12
41
51
72
104
3.0%
634
1 425
1 842
2 363
3 348
2.5% 1.0%
Non-OECD E. Europe/Eurasia
480
738
846
963
1 069
n.a.
583
651
712
794
0.8%
57
247
335
449
540
2.0%
China
14
27
59
104
115
2.9%
India
1
25
28
41
45
2.0%
38
204
324
483
999
4.8%
Russia Asia
Middle East Africa
23
133
197
286
452
3.5%
Latin America
36
104
139
182
287
3.1%
Brazil World European Union
1
7
11
17
38
5.2%
1 522
2 531
2 959
3 512
4 434
1.7%
n.a.
262
228
170
99
-3.4%
* Average annual rate of growth. Note: Historical data for world production differs from demand because of stock changes. 2. This section summarises the Reference Scenario projections of global natural gas production, which are described in detail in Chapter 12.
Chapter 4 - Natural gas market outlook
115
© OECD/IEA, 2008
Table 4.2
production costs, whether or not the gas is produced in association with oil. Iran, Qatar, Iraq and Saudi Arabia account for most of the increase. About 60% of incremental output is consumed locally, mainly in power stations, while the remainder is exported (see below). Most of the remaining increase in world output is provided by Africa and Eastern Europe/Eurasia (mainly Russia). The latter region remains the largest single producing region in 2030, just ahead of the Middle East. Production rises marginally in OECD North America over 2006-2030, mainly due to non-conventional supplies (notably shale gas), but falls heavily in Europe where North Sea reserves continue to dwindle. Australia accounts for virtually all of the projected increase in OECD Pacific output. Natural gas production by region in the Reference Scenario
Figure 4.5
2006
E. Europe/Eurasia
2030
Middle East OECD North America Asia Africa Latin America OECD Europe OECD Pacific 0
200
400
600
800
1 000
1 200
Billion cubic metres
Adequacy of gas-supply infrastructure and risks of shortages
The rigidity of gas-supply chains and the lengthening lead times in building physical infrastructure increase the risk that demand may, at times, outstrip available capacity. As the 2007 issue of the IEA’s Natural Gas Market Review put it “The world dodged a bullet” in the winter of 2006/2007: delays in bringing new capacity on stream did not lead to shortages in Europe and North America, mainly thanks to warmer-than-normal weather, which dampened demand (IEA, 2007). Yet concerns remain that investment in upstream facilities, LNG chains and long-distance pipelines, in the medium term, may fall short of that needed to meet growing demand, especially if power generators opt to build more gas-fired capacity than the high levels already projected (for example, if gas prices are lower than assumed). It is generally quicker to build gas-fired capacity than the infrastructure required to supply it. Limited spare capacity within the gas-supply system increases the risk of physical shortages occurring due to a sudden supply disruption or an exceptional surge in demand, for weather-related or other reasons.
116
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Box 4.1
Box 4.1
Adequacy of gas-supply infrastructure and risks of shortages (continued)
Recent supply problems in several markets have refocused attention on gassupply security. The loss of US gas-production capacity through hurricane damage in 2005 resulted in a surge in prices that choked off a substantial amount of industrial demand, rebalancing the market. The high degree of interconnectivity of the North American network enabled supplies to be redirected. In the event, no customer was forced to stop using gas. But other countries that have not liberalised their gas markets to the same degree cannot rely on market forces to bring demand back into balance with supply during a crisis. In the winter of “2005/2006”, cold weather and higher-than-usual use of gas for power generation led to a surge in demand in Italy. A lack of cross-border capacity and fixed prices to most customers resulted in physical shortages that could only be addressed through plant closures through administrative decree, the temporary relaxation of environmental standards to allow fuel oil to be used instead of gas and rationing. Commercial disputes with neighbouring countries have recently led to disruptions in Russian gas supplies to Europe. These incidents have highlighted the importance of giving full rein to markets in balancing supply and demand in the event of a crisis, and to the need for timely and adequate investment in transmission and storage capacity.
4
3. There has already been an increase in inter-regional swaps of LNG cargoes, particularly from the Atlantic to Pacific regions, facilitated by growing supplies and changing business models in the LNG industry.
Chapter 4 - Natural gas market outlook
117
© OECD/IEA, 2008
There are particular concerns about the adequacy of investment in LNG capacity in the medium term. An unprecedented major expansion is underway worldwide in regasification capacities, well in excess of LNG-production capacity. As a result, global regasification capacity is likely to be under-utilised, even though its availability will increase supply flexibility.3 But liquefaction capacity is not expected to increase as quickly. Although a significant amount of capacity is planned and proposed, many projects have yet to be formally sanctioned. Major delays and cost over-runs afflict many projects that have been given the green light, which has discouraged companies from proceeding to final investment decisions on other projects and has led to some cancellations. The dearth of investment decisions in new LNG projects since mid-2005 means that any new surge of investment is unlikely to result in additional capacity, beyond that due in service by 2012, being available before 2015, given the long lead times involved. Notwithstanding the massive expansion in LNG supply that will undoubtedly occur by 2012, the lag in LNG investment beyond 2012 raises questions about the availability of incremental LNG supply for both OECD and non-OECD importing countries. Shortfalls in the availability of LNG could push up prices and encourage the faster development of indigenous resources in importing regions.
118
bcm
- 97
Pacific
57
15
11
50
17
21
2
14
30
16
19
38
91
61
45
2
24
% of primary demand*
bcm
- 435
582
8
162
105
- 16
- 17
36
205
185
496
32
- 143
- 111
- 333
- 53
- 496
* Production for exporting regions. Note: Trade between WEO regions only. Positive figures denote exports; negative figures imports. Sources: IEA analysis and databases; Cedigaz (2008).
441
- 305
16
Latin America
European Union
99
World
55
Africa
-8
Middle East
-1
India
46
China
Asia
198
137
Russia
E. Europe/Eurasia
18
353
Non-OECD
OECD Oceania
- 115
- 241
Europe
OECD Asia
- 15
- 353
2006
72
17
4
57
22
29
14
8
29
19
21
47
97
61
54
6
30
% of primary demand*
2015
Net inter-regional natural gas trade in the Reference Scenario
North America
OECD
Table 4.3
© OECD/IEA, 2008
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
- 582
1 022
35
284
323
- 71
- 106
- 126
270
224
741
58
- 179
- 121
- 477
- 143
- 741
bcm
86
23
12
63
32
61
48
19
34
21
22
58
98
54
69
16
41
% of primary demand*
2030
Inter-regional trade Inter-regional natural gas trade (between major WEO regions) is projected to more than double over the projection period, from 441 bcm in 2006 to around 1 tcm in 2030 (Table 4.3). Trade rises much faster than demand, due to the pronounced geographical mismatch between resource location and demand. As a result, regional gas markets become more integrated as trade in LNG expands and new long-distance and undersea pipelines enable more gas to be traded between regions.
4
Imports rise in all the regions (with the exception of non-Russia Eurasia) that are currently net importers of gas, both in volume and as a share of total gas consumption (Table 4.3). The European Union sees the biggest increase in net import volumes, from 305 bcm in 2006 to about 580 bcm in 2030. Its reliance on imports rises from 57% of its total gas needs to 86% over that period. Most of the increase is met by Russia, Africa and the Middle East. North America, which is almost self-sufficient in gas at present, emerges as a major importing region during the projection period as demand outpaces production. Imports from Africa, Latin America and the Middle East, mostly into the United States, reach a combined 140 bcm in 2030. OECD Asia (Japan and Korea) is already highly reliant on imports to meet its gas needs. Its net imports, from non-OECD Asian countries, the Middle East, Australia and Russia (Sakhalin), rise by more than half between 2006 and 2030. Imports into China also increase sharply, covering just under half of the country’s gas needs in 2030. India’s imports also grow markedly, meeting more than 60% of total gas demand by 2030. Most of the additional exports over 2006-2030 come from the Middle East and Africa. Together, they account for 59% of total gas exports in 2030. Exports from Africa are projected to almost triple, from 99 bcm in 2006 to over 280 bcm in 2030, while those from the Middle East increase six-fold, from 55 bcm to about 320 bcm. African countries’ exports to Europe grow strongly. The Middle East increases its net exports to all the main consuming markets except China: Europe, North America, OECD Asia and India. Most of the additional exports from Africa and the Middle East are in the form of LNG. Russia and the Caspian/Central Asian countries collectively make up the other main exporting region. This region increases its exports to Europe by 14% and starts to export to China and OECD Asia. Russia’s total net exports rise from 198 bcm in 2006 to 270 bcm in 2030. Apart from the Sakhalin project, all exports from Russia and the Caspian/Central Asia are via pipeline.
4. The share of LNG in total international trade is significantly lower, as most intra-regional trade is by pipeline. LNG is generally the cheaper means of transport over long distances.
Chapter 4 - Natural gas market outlook
119
© OECD/IEA, 2008
Worldwide, LNG flows have doubled in the past decade and now meet 7% of total world demand for natural gas. The share of LNG in total trade between WEO regions continues to grow steadily through the Outlook period, from 52% in 2006 to 69% in 2030 (Figure 4.7).4 The volume of LNG trade reaches 340 bcm in 2015 and around 680 bcm in 2030 — up from only 201 bcm in 2006. LNG accounts for 80% of the increase in interregional trade.
120
2006
1
3
2030
35
OECD North America Latin America
11
6
4
12
93
36
70
7
1
106
4
Africa
India OECD Asia
1
61
Middle East
1
12
63
3
Eastern Europe/Eurasia
10
137
156
OECD Europe
261
88
2
8
47
58
Other Asia OECD Oceania
China
35
1
Main net inter-regional natural gas trade flows in the Reference Scenario, 2006 and 2030 (billion cubic metres per year)
The boundaries and names shown and the designations used on maps included in this publication do not imply official endorsement or acceptance by the IEA.
2
2
Figure 4.6
© OECD/IEA, 2008
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
23
17
3
Billion cubic metres
Figure 4.7
World inter-regional natural gas trade* by type in the Reference Scenario
1 200 1 000 800
80%
LNG
70%
Pipelines
60%
Share of LNG
4
50%
600
40% 30%
400
20% 200
10%
0
0% 2006
2015
2030
*Trade from major WEO regions, not including international trade within each region.
The pattern of world LNG trade is set to change markedly. Until now, LNG trade has been confined largely to the Asia-Pacific region, with gas sourced within Asia and in the Middle East. Although this market continues to expand, LNG demand from Atlantic Basin markets increases even more over the projection period, so that flows from Africa and the Middle East to Europe and North America quickly overtake those flowing eastwards. OPEC countries, especially in Africa and the Middle East, dominate the growth in supply of LNG through to 2030 (Figure 4.8). At the end of 2007, there were 24 liquefaction terminals in operation worldwide, with a total capacity of 256 bcm per year. An additional 146 bcm/year of capacity is under construction, which will take total capacity to around 400 bcm/year by 2012 (Table 4.4). Another 417 bcm/year of capacity is in the planning stage, though many of the planned projects have been on the table for several years. If only half of this planned capacity is built, total capacity would reach over 600 bcm/year.
Billion cubic metres
Figure 4.8
Inter-regional exports* of LNG by source in the Reference Scenario
700
OECD
600
Other non-OECD Africa
500
Middle East 400 300 200
0 2006
2015
2030
* Net exports of LNG from major WEO regions, not including international trade within each region.
Chapter 4 - Natural gas market outlook
121
© OECD/IEA, 2008
100
Table 4.4
Natural gas liquefaction capacity (billion cubic metres per year) Capacity end-2007* Capacity under construction** Projected capacity in 2012***
OECD Australia Norway United States Non-OECD Algeria Angola Brunei Egypt Equatorial Guinea Indonesia Libya Malaysia Nigeria Oman Peru Qatar Russia Trinidad and Tobago UAE Yemen World OPEC
23 21 – 2 233 28 – 10 16 1 38 1 31 25 15
18 13 5 – 128 6 7 – – 3 10 – 2 6 –
41 34 5 2 361 34 7 10 16 4 48 1 33 31 15
–
6
6
40 – 21 8 – 256 139
65 13 – – 9 146 94
105 13 21 8 9 402 233
122
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
* Capacity at liquefaction facilities commissioned in 2007 is included in the column under construction as they are not yet operated at full capacity. ** Includes ramping up capacity at liquefaction facilities commissioned in 2007 and capacity additions by debottlenecking. *** Planned and proposed capacity not under construction. Source: IEA (2008).
CHAPTER 5
COAL MARKET OUTLOOK H
I
G
H
L
I
G
H
T
S
Since 2000, global coal consumption has grown faster than any other fuel, despite higher prices, by 4.9% per year between 2000 and 2006. Most of this growth occurred in non-OECD countries. Coal demand is projected to grow by 2% per year over the Outlook period, faster than total energy demand. NonOECD countries account for 97% of the increase in global coal demand over the Outlook period, with China alone accounting for two-thirds of the increase and India for a further 19%. To meet growing demand, coal production is projected to rise by almost 60% between 2006 and 2030. Around 90% of this increase comes from non-OECD countries. China almost doubles its output, while India’s production more than doubles. Russian production jumps by nearly 75%, overtaking that of OECD Europe. Proven remaining reserves are more than adequate to meet the growth in coal use projected to 2030. The United States, Russia and China together account for around 60% of world reserves. However, large investments are needed in prospecting (to identify economically recoverable reserves) and in new mining projects. Global trade in coal between WEO regions rises from 613 million tonnes of coal equivalent today to around 980 Mtce by 2030. Despite a sharp decline in its coal exports, China remains a net exporter at present, but becomes a net importer in the near future, with net imports rising to 88 Mtce by 2030. India’s imports grow by 7% per year to 220 Mtce, overtaking Europe to become the world’s second-largest net importer after OECD Asia.
Carbon dioxide emissions from coal combustion are set to rise from 11.7 gigatonnes in 2006 to 18.6 Gt in 2030, driving up coal’s share of total emissions from 42% to 46%. Carbon capture and storage technology has the potential to reduce greatly CO2 emissions from coal use in the long term, but has only a small impact before 2030 in the Reference Scenario.
Chapter 5 - Coal market outlook
123
© OECD/IEA, 2008
Coal prices have kept pace with increases in oil and natural gas prices in recent years. The spot price of steam coal delivered to ports in Europe and Asia rose to above $100 per tonne in 2007 and continued to rise steeply in 2008. Coalsupply costs have also risen dramatically, due to sharp increases in the cost of materials, equipment, diesel, labour and shipping. Projected cumulative investment in coal-supply infrastructure totals about $730 billion between 2007 and 2030 (in year-2007 dollars), 91% of which is required for mines, and the rest for ports and shipping.
Demand Globally, coal consumption grew briskly between 2000 and 2006, at 4.9% per annum — a rate well above that of total energy demand, which grew by 2.6% per year. Coal demand grew faster even than the use of modern renewables, which rose at a rate of 3.1% per annum (Figure 5.1). Coal is the world’s second most important fuel, after oil, accounting for 26% of global energy demand. Global coal demand reached 4 362 million tonnes of coal equivalent (Mtce) in 2006. The world’s five largest coal consumers — China, the United States, India, Japan and Russia — account for 72% of total coal consumption. Between 2000 and 2006, coal demand in OECD member countries grew on average by just 0.6% per year, its share of global demand dropping to 37% in 2006, down from 48% in 2000. Preliminary 2007 data shows a renewed increase in OECD demand of 2.1% — twice the rate of growth of total primary energy.
Mtoe
Figure 5.1 800
Incremental world primary energy demand by fuel, 2000-2006
4.9%
% = average annual rate of growth
700 600 500 1.7%
400
2.4%
300 200
2.1%
100
1.3%
2.5%
3.1%
Nuclear
Hydro
Other renewables
0 Coal
Oil
Gas
Biomass
The detailed analysis in last year’s Outlook of China and India, two of the world’s largest coal consumers, led to a significant increase in projected global coal consumption compared with previous WEOs. In comparison, this year’s Outlook projects lower coal demand in OECD countries, due to recent delays in the construction of coal-fired power plants in the European Union and the United States, for environmental reasons, and because of a shift to natural gas and renewable energy in power generation. 124
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
The pace of coal growth is projected to slow over the Outlook period, to 3.1% per year on average through to 2015 and 1.3% thereafter to 2030 (Table 5.1). Despite this slowdown as users switch to less carbon-intensive alternatives, coal demand grows faster than overall energy demand. Coal consumption increases by 32% by 2015 and by 61% from 2006 to 2030, reaching 7 011 Mtce. The projected increase in coal consumption to 2030, at 2 649 Mtce, is almost equal to the combined coal consumption of all nonOECD countries today. With this strong growth, the share of coal in total energy supply rises from 26% in 2006 to 29% by around 2025, before levelling off through to 2030 as the growth in other fuels, particularly for power generation, eats into coal’s market share.
This is offset by slightly higher coal demand in non-OECD countries, driven by a greater emphasis there on economic factors and stronger growth in electricity demand. Overall, projected coal demand in 2030 is slightly lower than in last year’s Outlook, in line with recent rises in coal prices and the higher coal-price assumptions for the Outlook period. Table 5.1
World primary coal* demand in the Reference Scenario (million tonnes of coal equivalent) 1980
2000
2006
2015
2030
1 373
1 566
1 627
1 728
1 703
0.2%
North America
571
832
839
895
959
0.6%
United States
537
777
787
829
905
0.6%
Europe
657
467
472
491
418
–0.5%
Pacific
145
267
316
342
326
0.1%
OECD
Japan
2006-2030**
85
140
161
164
153
–0.2%
1 181
1 714
2 735
4 019
5 308
2.8%
517
295
307
356
386
1.0%
n.a.
158
152
201
233
1.8%
572
1 249
2 238
3 415
4 634
3.1%
China
446
899
1 734
2 712
3 487
3.0%
India
75
235
318
451
827
4.1%
2
12
13
20
36
4.4%
Africa
74
129
147
174
175
0.8%
Latin America
16
29
31
55
77
3.8%
2 554
3 279
4 362
5 746
7 011
2.0%
n.a.
459
463
460
372
–0.9%
Non-OECD E. Europe/Eurasia Russia Asia
Middle East
World*** European Union
5
* Includes hard coal (steam and coking coal), brown coal (lignite) and peat. ** Average annual rate of growth. *** Includes stock changes.
Switching from gas-fired to coal-fired power generation remains economically attractive in Russia, leading to a rise in coal demand of over 50% over the projection period, to reach 233 Mtce by 2030. China and India both see strong growth in coal consumption. China alone will account for two-thirds of the global increase Chapter 5 - Coal market outlook
125
© OECD/IEA, 2008
The OECD is projected to see an initial minor increase in coal consumption, from 1 627 Mtce in 2006 to about 1 730 Mtce in 2015, then a drop to just over 1 700 Mtce in 2030. Growth over the projection period averages only 0.2% per year, reflecting policies aimed at reducing carbon-dioxide (CO2) emissions. The share of coal in total OECD primary energy use drops from 21% today to 19% by 2030, driven mainly by a sharp drop in the share of coal as an input to power generation, from 41% today to 36% by 2030, as natural gas, biomass and waste, and other renewables all increase their share. In contrast with the situation in OECD Europe, the availability of low-cost, indigenous coal in OECD North America favours modest demand growth. The industrial sector in the OECD continues to become less coal intensive, with coal use actually falling by 0.6% per year over the projection period. Most coal use in OECD countries is still used in power generation: the sector’s share of total coal use is the same in 2030 as in 2006, at 81%.
in coal demand over 2006-2030, with India accounting for a further 19%. NonOECD countries in total account for 97% of the global increase in coal demand, mainly to meet rising electricity demand. China consumes 40% of the world’s coal today and this share rises to 50% by 2030. Its demand rises at an average rate of 3% per year. India sees the world’s second-fastest coal growth of 4.1% per year over the projection period. Although insignificant in global terms, coal demand in the Middle East is forecast to grow strongly by 4.4% per year as coal-fired power generation allows oil and gas to be used for more valuable applications or exported. Most of the coal consumed today worldwide is for power generation and this is not expected to change. This sector accounted for over two-thirds of global consumption in 2006, a share that remains broadly flat through to 2030 (Figure 5.2). Coal use in power generation increases at 2% per year. This overall average masks the intermediate trend of an increasing share of coal in power generation through to 2020, before falling back to 2006 levels by the end of the Outlook period. Industry remains the second largest consumer of coal, though growth in industrial coal use slows dramatically in the second half of the projection period, from 2.9% per year in 2006-2015 to 1.1% per year in 2015-2030. This is caused by the projected levelling out in non-OECD countries, particularly China, of iron production from blast furnaces, which is very coal intensive. Recycled scrap accounts for an increasing share of steel production, particularly from electric arc furnaces, indirectly creating demand for coal in the power sector. Coal use in the residential, services and agriculture sectors as a whole is projected to fall. Sectoral shares in coal demand by region in the Reference Scenario
Middle East
2006 2030
Power generation
OECD North America
2006 2030
Industry
Russia
2006 2030
Other sectors
OECD Europe
2006 2030
E. Europe/Eurasia
2006 2030
OECD Pacific
2006 2030
India
2006 2030
World
2006 2030
Latin America
2006 2030
Other Asia
2006 2030
Africa
2006 2030
China
2006 2030
0%
20%
40%
60%
80%
100%
The extent and effectiveness of policy action to reduce the carbon intensity of energy consumption, in order to mitigate climate change, remains the principal uncertainty for future trends in coal demand. Another factor that greatly affects coal’s prospects is the cost differential between coal and gas for power generation. While coal is the 126
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Figure 5.2
cheapest way to generate electricity in many regions (see Chapter 6), particularly where it is domestically available, the widespread introduction of a carbon value (such as under the EU Emissions Trading Scheme) could significantly reduce the attractiveness of investing in coal plants, resulting in lower coal demand than projected here. In 2006, CO2 emissions from coal use totalled 11.7 Gt, equal to 42% of global energy-related CO2 emissions (Figure 5.3). This share is ten percentage points higher than the share of coal in total primary fossil-energy use, due to the higher carbon intensity of coal compared to oil and gas. Global CO2 emissions from coal are projected to rise by 2% per year over the Outlook period, reaching 18.6 Gt in 2030. Coal’s share of CO2 emissions rises to 46%. The absolute increase in coal emissions is slightly larger than the level of current emissions from all fossil-fuel use in OECD North America. Coal use also results in emissions of other types of greenhouse gases (see Chapter 16), as well as other toxic and pollutant gases, which can have serious effects on the local environment and human health. Concerted efforts to combat climate change — including switching to lower-carbon fuels and the widespread deployment of carbon capture and storage technology — could lead to much lower emissions than projected in the Reference Scenario (see Part C).
Gigatonnes
Figure 5.3
5
World energy-related CO2 emissions by fuel in the Reference Scenario
20
50%
16
40%
12
30%
8
20%
4
10%
Coal Oil
0
Gas Share of coal in total emissions (right axis)
0% 1990
2000
2006
2015
2030
Coal is the most abundant and geographically dispersed fossil fuel. Proven reserves at the end of 2005 were 847 billion tonnes (WEC, 2007). The United States, Russia and China combined account for 61% of proven reserves (Figure 5.4). Australia, the world’s largest coal exporter, has 9% of the world’s proven reserves, while the second-largest exporter, Indonesia, has less than 1%. Around half of the world’s proven reserves are bituminous coal and anthracite, the grade of coal with the highest energy content. Current reserves are more than adequate to meet projected growth in coal demand through to 2030 in this Outlook. However the rapid increase in demand in recent years Chapter 5 - Coal market outlook
127
© OECD/IEA, 2008
Reserves and production
has seen the global reserves-to-production ratio fall sharply, from 188 years in 2002 to 144 years in 2005 (WEC, 2007 and 2004). This fall can be attributed to the lack of incentives to prove up reserves, rather than a lack of coal resources. Exploration activity is typically carried out by mining companies with short planning horizons, rather than by state-funded geological surveys. With no economic need to prove long-term reserves, the ratio of proven reserves to production is likely to fall further.
Figure 5.4
Proven coal reserves in leading producing countries, 2005 Bituminous including anthracite
Rest of world Kazakhstan
Sub-bituminous
Ukraine
Lignite
South Africa India Australia China Russia United States 0
50
100
150
200
250 Billion tonnes
Source: WEC (2007).
Between 2006 and 2030, world coal production is projected to rise by almost 60%, or over 2 610 Mtce (Table 5.2). The United States remains far and away the biggest producer in the OECD, its production rising by 19% between 2006 and 2030. Production in OECD Europe continues its steady decline. The removal of subsidies, notably in Germany, leads to the closure of high-cost mines. Coal output in OECD Pacific is projected to grow at 1.6% per year, almost all of the growth occurring in Australia.
The projected growth in coal production and use in China and India in the Reference Scenario hinges on the way those countries address several challenges, including local and global environmental impacts (see Part C). At the local level, the water and other infrastructure demands of coal mining in arid regions of China’s northwest will place enormous strains on a delicate ecosystem; in India, the loss of forests and villages, and the displacement of people, make any expansion of its largely open-cast coal industry politically challenging. The Chinese government has identified coal mining as a strategic sector and the industry is currently 128
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
In order to meet domestic coal demand, coal production in non-OECD countries grows significantly, expanding at 2.5% per year. Non-OECD countries account for 91% of the increase in world coal production over the Outlook period. China almost doubles its current coal production, though growth slows after 2015. In India, output increases by a factor of 2.1. Robust growth in production in Russia sees it overtake OECD Europe around 2010 and continue to expand thereafter.
profitable with ample funds for investment. Proving coal reserves and installing the transport infrastructure to move large volumes of coal over ever-longer distances are practical challenges that China has already demonstrated it can surmount with the development of coalfields in northern Shanxi, Shaanxi and Inner Mongolia. The investment requirements, though large in absolute terms, are small relative to China’s whole energy sector. In India, the projected expansion of output assumes that the coal sector, dominated by Coal India, becomes more competitive. Table 5.2
World coal production in the Reference Scenario (million tonnes of coal equivalent)
OECD
5
1980
2000
2006
2015
2030
2006-2030*
1 378
1 384
1 446
1 566
1 684
0.6%
North America
672
835
878
962
1 048
0.7%
United States
640
778
824
894
981
0.7%
Europe
603
306
273
246
208
–1.1%
Pacific
104
243
294
358
427
1.6%
Oceania Non-OECD
77
238
293
358
427
1.6%
1 196
1 792
2 950
4 180
5 327
2.5% 1.3%
E. Europe/Eurasia
519
306
357
443
481
n.a.
167
205
295
354
2.3%
568
1 250
2 316
3 367
4 435
2.7%
China
444
928
1 763
2 647
3 399
2.8%
India
77
209
283
352
607
3.2%
1
1
1
2
3
2.2%
100
187
203
257
271
1.2%
Russia Asia
Middle East Africa Latin America World
9
48
73
110
137
2.7%
2 574
3 176
4 396
5 746
7 011
2.0%
n.a.
306
273
232
180
–1.7%
European Union
* Average annual rate of growth.
World coal production by type in the Reference Scenario
8 000
Brown coal and peat
7 000
Coking coal
6 000
Steam coal
5 000 4 000 3 000 2 000 1 000 0 1990
Chapter 5 - Coal market outlook
2006
2015
2030
129
© OECD/IEA, 2008
Million tonnes of coal equivalent
Figure 5.5
Steam coal, used mainly in power generation, will remain the main type of coal produced worldwide. Output is projected to grow by 2.2% per year and its share of total coal output to rise from 77% in 2006 to 81% in 2030, driven by power generation in Asia. The projected growth of coking coal, at 1.3% per year, is roughly in line with the 1.7% per year coal growth projected in the iron and steel sector (the main consumer) over the Outlook period. Use of brown coal and peat also grows slowly, by 1.2% per year. Box 5.1
Alternative energy supplies: underground coal gasification and coal-mine methane1
Underground coal gasification (UCG) offers the prospect of bringing more of the world’s coal resources to market, not as a solid fuel, but as a clean gaseous fuel. Conventional underground coal mining becomes more difficult and more expensive as mining depth increases. Mining below 1 000 metres is generally not economic. Yet 60% of China’s coal resources, for example, lie below 1 000 metres (Pan, 2005). In addition, coal lying offshore is generally not practical to mine, but could be exploited using UCG. Estimates of the coal resources suitable for UCG vary widely, but a conservative estimate suggests that they could yield the equivalent of 146 trillion cubic metres (WEC, 2007) — similar in magnitude to the world’s current conventional natural gas reserves.
The technical challenge with UCG is to control the hot cavity so that it moves along within the coal seam and converts as much coal as possible into useful fuel gas. Developments in directional drilling over the last decade mean that this challenge can be overcome and an efficient drilling pattern established to recover the maximum amount of energy. The seismic and hydrogeological skills needed to design UCG projects are to be found in the oil and gas industry, though they are in short supply. Environmental concerns relate mainly to the fate of hazardous by-products. In well-designed schemes, these are handled in surface treatment plants or remain immobile within the gasified coal seam. The prospects for UCG are improving, as energy prices rise and technology develops. But until it is demonstrated on a commercial scale, it remains only an interesting prospect: it is not assumed to have a material impact on energy supplies over the Outlook period. 1. Coalbed methane is discussed in Chapter 12.
130
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
UCG draws on technologies, developed in the oil and gas industry, to drill into coal seams, introduce an oxidant to combust some coal, and create a hot cavity in which more coal is gasified with steam. The resulting gas (mainly hydrogen and carbon monoxide, with a smaller quantity of methane) is drawn off to fuel a gas turbine for large-scale power generation or as feedstock for chemical or liquid fuel production. UCG was practised on a large scale in the former Soviet Union from the early 1960s, with one site in Uzbekistan remaining in production to supply a 400 MW power plant. Subsequent trials in Europe and the United States, and more recently in China, Australia and South Africa, have not yet led to any commercial UCG projects. But the prospects for UCG are tempting enough to have attracted some private investment, notably by Linc Energy of Australia.
Coal-mine methane (CMM) involves capturing methane released from the coal seam and surrounding strata before or during coal mining. In underground mines, ventilation must be sufficient to keep the methane concentration low enough to avoid explosions. Since the build up of methane can slow mining, it is common practice to pre-drill coal seams and draw off methane before mining. The purity of the captured methane, typically of the order of 35% to 75% (Creedy and Tilley, 2003), means it can be used on-site for power generation, sold for local use or purified for sale as pipeline gas, although the latter option is only viable for larger projects. It is also possible to collect methane from the “gob” areas left behind after coal extraction. The methane content of coal can reach 25 cubic metres per tonne, increasing with depth and coal rank. Although the average value might be around 5 m3 per tonne, methane release per tonne of coal mined is much larger, since more coal is disturbed during mining than is actually extracted.2An estimated 28 bcm of methane was released in 2005 during coal mining worldwide, or 6% of total global methane emissions from human activity (EPA, 2006).
5
In developing countries, CMM projects can benefit from Clean Development Mechanism credits under the Kyoto Protocol. This has led to a wave of new projects in China, where the value of avoided methane emissions can be significant. Methane has a global warming potential at least 21 times that of CO2, so even simple projects that flare rather than vent methane can have a significant environmental and financial value.
Inter-regional trade3 Global trade in hard coal between WEO regions in 2006 amounted to 613 Mtce, or 14% of total hard coal output (Table 5.3). Trade is projected to grow at 2.0% per year through to 2030, reaching almost 980 Mtce. Trade grows quickest in the period until 2015, at 3.1% per year. Trade grows slightly slower than demand, as most coal continues to be consumed within the region in which it is produced. As a result, the share of global hard coal output that is traded between WEO regions remains flat at 14%. Major demand centres in non-OECD countries, China and India, see rapid growth in imports over the projection period (Figure 5.6). China was a net exporter with a peak of 87 Mtce in 2001, but net exports have declined and China is expected to become a net importer in the near future. Total net imports are projected to reach 88 Mtce by 2030. India’s net imports also grow strongly, by 7% per year, resulting in India overtaking OECD Europe to become the world’s second-largest net importing region after OECD Asia. Europe’s imports rise through to the middle of the projection period, but then fall back by 2030 to a level close to that of today, as coal demand for power generation drops.
2. IPCC (2006) guidelines suggest a range of 10-25 m3 per tonne of coal mined from underground when reporting emissions, before accounting for any methane captured. 3. All the trade figures cited in this section exclude brown coal and peat.
Chapter 5 - Coal market outlook
131
© OECD/IEA, 2008
Russia is currently the world’s third-largest net-exporting country. At the higher prices assumed in this Outlook compared with last year, Russian exports are very competitive,
despite the long rail-transport distances from the Kuzbass coalfields of central Russia to ports serving the Atlantic and Pacific markets. There are plans to expand port capacities, with exports to the Far East expected to grow quickly from a low base. Production can easily be expanded to meet both rising domestic demand and export growth. Table 5.3
Net inter-regional hard coal* trade in the Reference Scenario (million tonnes of coal equivalent) 1980
OECD
2000
2006
2015
2030
– 19
– 122
– 203
– 162
80
45
15
67
90
Europe
– 67
– 150
– 202
– 245
– 209
Pacific
– 32
– 18
– 16
16
101
Asia
– 73
– 193
– 232
– 256
– 246
North America
– 19
Oceania
41
175
216
271
347
Non-OECD
16
119
212
162
19
0
13
52
88
96
n.a.
10
53
94
120
E. Europe/Eurasia Russia Asia China India
–2
35
73
– 47
– 199
5
58
34
– 65
– 88
0
– 20
– 42
– 98
– 220
–1
– 10
– 12
– 18
– 34
Africa
26
60
56
83
95
Latin America
–8
20
42
56
60
Middle East
World
154
456
613
804
979
European Union
n.a.
– 142
– 191
– 229
– 191
* Steam and coking coal, and coke. Note: Trade between WEO regions only. Positive figures denote exports; negative figures imports. Inter-regional trade between WEO regions differs from total international trade, which includes trade within regions.
Indonesia’s coal sector has repeatedly shown its ability to meet rising international demand for steam coal. Production is expected to continue to increase with modest investment expected in rail infrastructure to access coal resources lying beyond what 132
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Australia remains the world’s largest coal-exporting country. Australia, with ample lowcost reserves, has the potential to meet much of the increased demand for internationally traded coal. Infrastructure development will be crucial to realise this potential. Recent announcements of port expansion projects in Queensland and New South Wales are a step in this direction and other projects are likely to follow, including those that boost rail capacity. Given their higher value, coking-coal exports will be particularly important, especially if infrastructure development does not keep pace because of local opposition.
it is practical to transport with the current combination of trucks and river barges. Nonetheless, net exports grow only modestly, as rapid growth in domestic demand absorbs most of the projected increase in production.
Figure 5.6
Net inter-regional hard coal* trade in the Reference Scenario 2006
OECD Oceania Russia Other Asia Africa OECD North America Latin America Other E. Europe/Eurasia Middle East China OECD Europe India OECD Asia –300
2030
–200
–100
0
100
5
200 300 400 Million tonnes of coal equivalent
* Steam and coking coal, and coke.
OECD North America remains an important coal exporter throughout the projection period. US coal exports have dwindled since the 1990s, but are now growing strongly in response to very high coking-coal prices and European demand for steam coal. The United States is expected to remain a swing supplier in the international market, helping to meet demand and thus dampening prices. Western Canadian coking-coal producers are enjoying a revival and have a number of new projects planned. Long rail-transport distances from the coalfields to ports mean that Canadian coking coal is likely to remain less attractive economically than Australian coking coal.
Colombia is well placed to supply customers in the Atlantic steam coal market, filling demand left unmet by South African producers. It appears likely that the major private coal companies will work closely with government to ensure that new rail Chapter 5 - Coal market outlook
133
© OECD/IEA, 2008
Africa remains an important net exporter over the projection period. South Africa is expanding its principal coal-export terminal at Richards Bay to 91 million tonnes (Mt) and producers are optimistic about boosting exports, despite a recent period of limited growth and even decline. Fully utilising this capacity hinges on investment in rail infrastructure, to expand capacity and to extend the network to new producing regions. Domestic coal demand is buoyant with the re-opening by Eskom in 2008 of coal-fired power stations to meet strong electricity demand growth. Consequently, we project only a modest rise in exports to 2030. Mozambique becomes a coking-coal exporter by 2010 with the launch of a project by the Brazilian company Vale. This trend towards upstream coal-supply investment by coal users is also seen amongst some European power utilities. China has also followed this route, with its largest coal producer, Shenhua, now investing in Indonesia.
infrastructure and roads are built to allow exports to grow to well over 100 Mt per year. Venezuela has the potential to export considerably more than its current 8 Mt per year. A deepwater port, linked to Vale’s Socuy project, could see exports grow to 25 Mt. Latin American exports are projected to increase by over 40% between today and 2030. China’s net position in the international coal market is highly uncertain. It has the infrastructure and reserves to export hundreds of millions of tonnes per year and exports did grow sharply during the late 1990s, through to a peak in 2003 of 101 Mtce (net of imports, they peaked at 87 Mtce in 2001). The country is well placed to supply Japanese and South Korean markets. However, to ensure that strong domestic demand is met, the government has imposed export restrictions in the form of taxes and annual quotas on the four companies holding export licences. As China continues its market reforms, the domestic price of coal should better reflect its true value in the global market and allow China’s balance of coal imports and exports to reach a natural level. Vietnam has been an important supplier to China in recent years, but rising domestic demand and limited resources will see this trend reverse over the Outlook period. Mongolia is already developing mines and infrastructure to meet demand from China. We expect Mongolian coal exports to China to increase rapidly.
Coal prices and supply costs
While oil and natural gas prices influence coal prices through contractual linkages and opportunities for fuel switching, notably in power generation, there are many other factors which have contributed to higher prices. Supply has been tight during a period of unprecedented demand growth. Unforeseen events in late 2007 and early 2008 have added to this tightness: flooding of mines in Australia, electricity shortages in South Africa and a temporary halt to all Chinese coal exports as the Chinese government sought to ensure that domestic demand was met. Strong demand for steel production and power generation has seen the world’s bulk carrier shipping fleet struggle to meet demand to transport iron ore, coking coal, coke and steam coal. Shipping rates rose steeply during 2007 and have been very volatile during 2008, touching an historic high in May. However, freight rates may now fall as the number of new vessels earmarked for commissioning between 2008 and 2013 exceeds projected demand (Rogers, 2007). High coking-coal prices have had a knock-on impact on the price of good-quality steam coal, which can be used as a substitute in some cases. These factors, coupled with rising production costs, have 134
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Coal prices have kept pace with the surge in international oil and gas prices (Figure 5.7), although on an energy basis, coal remains the cheapest fossil fuel. For example, the average price of steam coal imported into the European Union in 2007 was $3 per MBtu, compared with $8.80 per MBtu for imported high-sulphur fuel oil, $7.30 per MBtu for natural gas imported by pipeline and $6.50 per MBtu for liquefied natural gas. During 2007 and 2008, coal prices repeatedly reached historic highs. Asian and European marker prices doubled in 2007 to above $100 per tonne CIF and have risen further during the first half of 2008, to above $200 per tonne in Europe.
contributed to higher prices in real terms. In fact, the increase has been bigger than that experienced during the mid-1970s to early 1980s, when coal prices last peaked.
Index (January 2003 = 100)
Figure 5.7
Index of selected fuel prices Oil - NYMEX WTI
600
Diesel - Northwest Europe
500
Gas - ICE NBP* 400
5
Steam coal - Asian CIF
300 200 100 0 Jan-03
Jan-04
Jan-05
Jan-06
Jan-07
Jan-08
Aug-08
* InterContinental Exchange National Balancing Point. Sources: Oil, diesel and gas: Platts. Steam coal: The McCloskey Coal Group Ltd, McCloskey’s Coal Report.
The reasons for cost increases in recent years have included escalating material costs (including tyres, explosives, spare parts and steel products), rising labour costs, depletion of the lowest-cost reserves, increased stripping ratios at open-cast mines and, in some cases, longer transport distances (Meister, 2008). Labour costs have risen everywhere, by as little as 5% in the United States to more than 40% in South Africa between 2003 and 2008. Rising oil prices have had a particular impact on open-cast mining, as widely practiced in Australia, Colombia, Indonesia, India, South Africa, Venezuela and the United States. Diesel fuel is essential for open-cast mining, particularly operations that rely heavily on mechanical shovels for extraction and trucks for coal haulage. While open-cast mining costs are generally lower than deep mining, diesel consumption can account for 20% of the total cash cost at open-cast operations. The cost of diesel to mining companies has risen by 100% to 500% in most coal-exporting countries since 2003, with Indonesia at the upper end of this range. Some of the increase in non-OECD countries is due to the removal of subsidies on diesel. Chapter 5 - Coal market outlook
135
© OECD/IEA, 2008
The steam-coal supply-cost curve (Figure 5.8) has shifted up sharply in recent years. Although the curve shows average cash costs (excluding capital depreciation costs) for each of the major coal-exporting countries, individual mine costs can vary widely. For example, Australian exporters operate across a broad range, with costs as low as $20 to more than $50 per tonne. At the high end, steam coal may be sold as a by-product of high-cost deep mines producing coking coal as a primary product. Indonesia, South Africa, Venezuela and Colombia all export coal produced at an average cost of around $30 per tonne, but individual mines range from $20 to $40 per tonne.
Dollars per tonne
Figure 5.8
Coal supply cash-cost curve for internationally traded steam coal for 2007 and selected average FOB prices for 2007 and first-half 2008
140
Vietnam
120
Venezuela Indonesia
100
Colombia
80
South Africa
60
China Queensland
40
New South Wales
20
Russia United States
0 5
55
105
155
205
255
305
355
405
455
505 555 580 Billion tonnes
Note: Adjusted for 6 000 kcal/kg. The lower band for prices represents 2007 and the upper band first-half 2008. FOB is free on board Source: IEA Clean Coal Centre.
Despite these significant cost increases, a large margin remains between coal prices and costs, which is attracting new investment into the industry and is stimulating considerable merger and acquisition activity, notably BHP Billiton’s proposed acquisition of Rio Tinto. This should help overcome one of the factors behind tight coal supply; but new capacity will take several years to come on line and, with demand growth remaining strong, prices are expected to remain high in the short term. Further concentration of the industry, particularly amongst suppliers of coking coal, works against a competitive investment climate and explains why, while some coal users are investing upstream, others are turning to competition regulators.
The overall investment needs of the coal industry were examined in detail in WEO-2003, when it was estimated that $400 billion (year-2000 dollars) would be required over the period 2001-2030. Since then, capital costs have increased significantly, with rising steel prices and engineering costs. In WEO-2003, the average mine was assumed to require investment of $34 per tonne of annual capacity (in year-2000 dollars). In this Outlook, we assume a $50 per tonne capacityinvestment cost (in year-2007 dollars) over the projection period. Some projects announced in 2008 involve much higher costs, of over $80 per tonne, but over the Outlook period we expect real investment costs to moderate. These cost estimates have been applied in order to calculate the total investment required in new and replacement capacity over the Outlook period to meet demand in the Reference Scenario. Investment of $1.20 per tonne produced (in year-2007 dollars) is added 136
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Coal investment
in order to provide for maintaining existing mines during their economic lives, to replace machinery and equipment, and to access new reserves. Increases in port capacity and shipping investment costs have also been taken into account. Structure of coal-supply costs for major exporting countries
Dollars per tonne
Figure 5.9 50
Port loading Inland transport
40
Taxes and royalties
5
Free at mine cost
30 20 10 0 Indonesia
Colombia
South Africa
China
Australia
Russia
Note: Adjusted for 6 000 kcal/kg. Source: IEA Clean Coal Centre.
In total, investment in the coal industry over 2007-2030 is now projected to amount to $728 billion (in year-2007 dollars). Some 91% of this is for mines, and the rest for ports and shipping (Figure 5.10). Investment in new capacity includes both surface and underground mine development, new machinery and infrastructure. Projected coal investment is 22% higher than in last year’s Outlook, reflecting the recent surge in costs as described above. China alone accounts for 44% of this investment. Despite these upwards revisions, coal investment accounts for only 3% of global energy-supply investment requirements. Cumulative investment in coal-supply infrastructure by region in the Reference Scenario, 2007-2030
Figure 5.10
Mining
China
Ports
OECD North America India E. Europe/Eurasia OECD Pacific OECD Europe Other Asia
Shipping 0
50
100
150
200
250
300
350
Billion dollars (2007)
Chapter 5 - Coal market outlook
137
© OECD/IEA, 2008
Rest of world
© OECD/IEA, 2008
CHAPTER 6
POWER SECTOR OUTLOOK H
I
G
H
L
I
G
H
T
S
Most of the projected growth in electricity demand occurs outside the OECD. In the OECD, electricity demand is projected to rise by just 1.1% per year on average, increasing by less than a third between 2006 and 2030. By contrast, demand in nonOECD countries grows by 146%, at an average annual rate of 3.8%. Globally, coal remains the main fuel for power generation throughout the period to 2030. Its share in total generation increases from 41% to 44%. Most of the growth in coal-fired generation occurs in non-OECD countries, which are projected to generate over two-thirds of all coal-fired electricity, compared with less than half now. The overall efficiency of coal-fired generation is projected to increase from 34% in 2006 to 36% in 2015 and to 38% in 2030. Plants with carbon capture and storage make only a minor contribution to generation in 2030. Stronger policies would need to be put in place — especially a more generally applied carbon constraint — for a significant amount of CCS capacity to be commissioned before 2030. The share of natural gas in the power-generation fuel mix falls, as a result of higher prices. Nuclear power loses market share, dropping from 15% in 2006 to 13% by 2015 and to 10% by 2030. The share of renewables rises considerably: it grows from 18% of total electricity generation in 2006 to 20% in 2015 and 23% by 2030. Total capacity additions — including replacement and expansion — amount to about 4 530 GW over the Outlook period, of which about 1 690 GW is added in the period to 2015. These additions average around 190 GW per year through to 2030 — about the same as over the past few years. Total cumulative investment in the power sector over the period 2007 to 2030 is projected at $13.6 trillion in year-2007 dollars, just over half of all energy-sector investment. The cost of building power stations has increased sharply over the past few years, notably in OECD countries, because of increases in the cost of materials, high energy prices, rising labour costs and supply-chain constraints. Coal and gas prices are assumed to remain high. As a result, the electricity produced in new plants that come on line in the next few years is increasingly costly, pushing up end-user electricity prices.
Chapter 6 - Power sector outlook
139
© OECD/IEA, 2008
World electricity demand in the Reference Scenario is projected to grow at an annual rate of 3.2% in the period 2006 to 2015, slowing to 2% per year on average in 2015-2030. This reflects a shift in the economies of non-OECD countries away from energy-intensive heavy manufacturing towards lighter industries and services, as well as saturation effects in the OECD and some emerging economies.
Electricity demand Global demand for electricity grew by nearly a quarter between 2000 and 2006, with nearly three-quarters of the increase coming from non-OECD countries (Table 6.1).1 Demand in OECD countries increased by just 10% during that period, while the increase in non-OECD countries exceeded 50%. China alone saw a doubling of its electricity demand, contributing over 40% of the total increase worldwide. Overall, world electricity demand grew by 3.6% per year in the period 2000 to 2006, following closely the growth in gross domestic product. Electricity demand is projected to grow at an annual rate of 3.2% in the period 2006 to 2015, slowing to 2% per year on average in the period 2015 to 2030. The projected slowdown in demand reflects a shift in the economies of non-OECD countries away from energy-intensive heavy manufacturing towards lighter industries and services, as well as efficiency improvements and saturation effects in the OECD and some emerging economies. Table 6.1
World final electricity consumption by region in the Reference Scenario (TWh) 2015
2030
2006-2030*
OECD
4 740
1980
8 251
9 035
10 177
11 843
1.1%
North America
2 386
4 144
4 413
4 870
5 774
1.1%
United States
2 026
3 500
3 723
4 045
4 723
1.0%
Europe
1 709
2 694
3 022
3 469
3 980
1.2%
Pacific
645
1 413
1 601
1 837
2 089
1.1%
Japan
513
944
981
1 061
1 162
0.7%
Non-OECD
2 059
4 390
6 630
10 580
16 298
3.8%
E. Europe/Eurasia
1 101
1 023
1 165
1 514
1 860
2.0%
n.a.
609
682
912
1 081
1.9%
477
2 023
3 669
6 574
10 589
4.5%
China
259
1 081
2 358
4 554
6 958
4.6%
India
90
369
506
893
1 935
5.7%
75
371
539
793
1 353
3.9%
Africa
158
346
479
667
997
3.1%
Latin America
248
627
777
1 032
1 498
2.8%
Russia Asia
Middle East
2000
2006
Brazil
119
319
375
478
651
2.3%
World
6 799
12 641
15 665
20 757
28 141
2.5%
n.a.
2 518
2 814
3 186
3 612
1.0%
European Union
1. Electricity demand refers to end-use consumption of electricity in industry, buildings, transport and agriculture. It is less than electricity generation because of losses in transmission and the own use of electricity at power plants.
140
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
* Average annual rate of growth.
World demand for electricity grows from 15 665 TWh in 2006 to about 20 760 TWh in 2015 and to about 28 140 TWh in 2030. The level in 2030 is lower than projected in last year’s WEO, mainly because of slower economic-growth expectations. Higher fossil-fuel prices in this year’s WEO also depress demand for electricity, as they lead to higher electricity prices. But this effect is partially offset by the improved competitiveness of electricity in end uses against direct use of fossil fuels for space and water heating, as electricity prices rise less rapidly in percentage terms. Non-OECD countries account for most of the projected growth in world electricity demand. In the OECD, electricity demand is projected to rise by just 1.1% per year on average, increasing by around 30% between 2006 and 2030. By contrast, demand in non-OECD countries grows by 146% over the same period, averaging 3.8% per year. Demand grows fastest in Asia (Figure 6.1). China’s electricity demand has been growing at an average annual rate of 14% since 2000. This high rate is expected to slow over time. It drops to 7.6% per year in the period to 2015 and averages 4.6% per year over the entire Outlook period. The projected slowdown results primarily from a shift in the economic structure from heavy industry towards less energy-intensive lighter industry and services. Even at lower growth rates, China soon becomes the biggest electricity consumer in the world, leaving behind the United States and the European Union. In Russia, electricity demand is likely to grow at a rate slightly above that of the OECD; growth in demand there is driven by increased electrification in industry and rising household income, which boosts the use of appliances.
Figure 6.1
6
Electricity demand growth rates by region in the Reference Scenario
8%
2006-2015
7%
2006-2030
6% 5% 4% 3% 2% 1% 0% Russia
Brazil
Africa
Middle East
China
Indonesia
India
Globally, industrial demand for electricity grows faster than demand for electricity by households and the services sector, driven primarily by rapid industrialisation in non-OECD countries. In OECD countries, demand for electricity in industry grows at a modest 0.6% per year, reflecting a long-term trend toward lighter manufacturing. Chapter 6 - Power sector outlook
141
© OECD/IEA, 2008
OECD
Despite strong growth in electricity demand in non-OECD countries, a large number of people living there are not expected to have access to electricity even in 2030. India and Africa have the highest number of people in this category.2 Per-capita electricity consumption in non-OECD countries doubles by 2030, rising to almost 2 400 kWh per person, but it is still well below the current OECD average of 7 641 kWh per person. There are large differences between countries: in China, per-capita electricity demand rises strongly, from 1 788 kWh in 2006 to 4 776 kWh by 2030, while in Africa, it rises modestly from 518 kWh now to 671 kWh (Figure 6.2).
kWh
Figure 6.2
Per-capita electricity demand by selected region in the Reference Scenario
10 000
2006 2030
8 000 6 000 4 000 2 000 0 OECD
Russia
Middle East
China
Brazil
India
Indonesia
Africa
Electricity supply Outlook for electricity generation In the Reference Scenario, global electricity generation rises from 18 921 TWh in 2006 to 24 975 TWh by 2015 and to 33 265 TWh by 2030.3 The largest increase is in non-OECD countries where electricity generation in aggregate matches that of the OECD countries soon after 2010 and by 2030 is almost 50% higher.
2. Chapter 15 discusses electricity access in resource-rich countries of sub-Saharan Africa. 3. Electricity generation includes final consumption of electricity, network losses and own use of electricity at power plants. Electricity generation figures exceed those for demand (final consumption) because of network losses and own use of electricity at power plants.
142
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Coal remains the main fuel for power generation worldwide throughout the period to 2030. On the back of strong growth in non-OECD countries, its share increases from 41% to 44% (Figure 6.3). The share of natural gas in total generation falls slightly, as a result of higher prices. The share of oil drops to about 2% by 2030, as high oil prices make oil burning extremely expensive. Nuclear power, constrained by the assumption
of unchanged government policies, also loses market share, which drops from 15% in 2006 to 13% by 2015 and further to 10% by 2030, as nuclear power capacity does not increase as rapidly as demand for electricity. Nuclear electricity generation increases from 2 793 TWh in 2006 to almost 3 460 TWh in 2030. The share of renewables rises considerably, from 18% in 2006 to 20% in 2015 and 23% in 2030.
TWh
Figure 6.3
World electricity generation by fuel in the Reference Scenario
16 000
Coal
14 000
Oil
12 000
Gas
10 000
Nuclear
6
Hydro
8 000
Biomass
6 000
Wind
4 000
Rest of renewables
2 000 0 2006
2030
Over 600 GW of power-generation capacity is currently under construction around the world, which is expected to be operational before 2015. Three-quarters of this new capacity is being built outside the OECD (Figure 6.4). About 200 GW is reported to be under construction in China, although this figure may be largely underestimated. These data provide useful insights into the future patterns of supply. Large amounts of coal-fired and hydropower capacity are set to come on line over the next few years. A smaller increase in nuclear power is in prospect. Gas-fired plants — combined-cycle gas turbines (CCGTs) and open-cycle turbines (OCTs) — which have short construction Chapter 6 - Power sector outlook
143
© OECD/IEA, 2008
Most electricity is projected to continue to come from centralised power stations, although there is a growing trend towards greater use of distributed generation: as increasing demand for reliable power supply and government policies to encourage the use of combined heat and power (CHP) and renewable energy encourage this solution. Good opportunities exist to provide heat and power for remote rural and urban communities. The use of CHP is projected to increase, based on biomass, natural gas and coal. Generating electricity and heat simultaneously offers an important opportunity to increase the overall efficiency of electricity and heat generation, and to reduce the environmental footprint. Providing for heat capture entails only a modest increase in investment costs and a slight loss of efficiency in electricity generation. Hence, CHP can be very profitable. The challenge is to match the heat supply with the demands of the industrial process. For district-heating systems, demand for heat can be spread across larger areas, but this requires substantial investment in heatdistribution infrastructure.
times (two to three years for a CCGT and one to two years for an OCT) — and landbased wind turbines, which have even shorter construction times (one to two years) may change the picture suggested by Figure 6.4 by 2015. In the OECD, there is more gas-fired capacity under construction than coal. Most of this is being built in Europe, because of tightening CO2 regulations and because gas is valued for its flexibility.
GW
Figure 6.4
Power-generation capacity under construction worldwide
250
Total = 613 GW
Non-OECD OECD
200 150 100 50 0
Coal
Gas
Oil
Nuclear
Hydro
Wind
Rest of renewables
Note: Includes power plants considered as under construction in 2007. Source: Platt’s World Electric Power Plants Database, January 2008 version.
Electricity generation from coal-fired power plants worldwide reached 7 756 TWh in 2006, over 40% of the total. In recent years, most of the growth in coal-fired generation has occurred in non-OECD countries, notably in China, where it more than doubled between 2000 and 2006. Since the beginning of market liberalisation in the 1990s, most new capacity in OECD has been gas-fired, although a large amount of coal-fired capacity was added in the OECD Pacific region. New coal-fired plants with a capacity of 28 GW have come on line since 2000; 18 GW of this capacity was in OECD Pacific, 6 GW in OECD Europe and 4 GW in OECD North America. Sharp increases in gas prices have made the economics of new coal-fired power plants more favourable and many power companies have announced plans to build additional coal-fired capacity. Difficulties have arisen in many cases, however, in moving from the planning phase to construction and some projects have been abandoned, particularly in the United States and Europe, because of local opposition driven by environmental concerns. Some projects have also been abandoned because of rising construction costs and because of regulatory uncertainty about climate policies. A total of 31 GW is now under construction in OECD countries (Figure 6.5). 144
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Trends in coal-fired generation
GW
Figure 6.5
Coal-fired capacity under construction in OECD countries
16
Mexico Canada United States
12
Turkey Poland
8
Italy Germany
4
Korea
6
Japan Australia
0 OECD North America
OECD Europe
OECD Pacific
Note: Includes power plants considered as under construction in 2007. Source: Platt’s World Electric Power Plants Database, January 2008 version.
Globally, coal-based electricity is projected to rise to 11 100 TWh by 2015 and to almost 14 600 TWh by 2030, giving rise to significant increases in associated CO2 emissions (see Chapter 16). Most of the growth takes place in non-OECD countries, which generate over two-thirds of all coal-fired electricity in 2030, compared with less than half now. Coal’s share of total electricity generation increases from 41% in 2006 to 44% in 2015 and remains broadly at this level through to 2030. More than half of coalfired capacity now under construction is based on subcritical technology, while the remainder is mainly supercritical. Over the Outlook period, supercritical technology is expected to become the norm, first in the OECD. Ultrasupercritical technology is projected to become more widespread in the OECD after 2020.
The average efficiency of coal-fired generation is projected to increase from 34% in 2006 to 36% in 2015 and to 38% in 2030 (Figure 6.6). In the OECD, efficiency increases from 37% to 41%. Although electricity demand there does not increase much, there is a significant opportunity to improve efficiency as old coal-fired power plants are retired. In non-OECD countries, efficiency improves from 31% to 37% because new coal-fired power plants built to meet rapidly growing electricity demand are expected to be far more efficient than existing ones. Chapter 6 - Power sector outlook
145
© OECD/IEA, 2008
Integrated gasification combined-cycle (IGCC) technology is expected to become more widespread after 2020. A small number of plants currently operating were initially built with public funding for demonstration purposes, with the best one achieving 42% electric efficiency (IEA, 2008a). There are plans to build IGCC plants in a number of countries, including the United States, the United Kingdom, Australia, Canada, Germany, the Netherlands, Japan, Poland, Czech Republic, Indonesia, India and China, though not all of them are expected to go ahead. The United States has the largest number of projects but the announced construction costs are high. Under the assumption that construction costs fall, improving the economics of IGCC, a total of 105 GW is projected to be built by 2030.
Figure 6.6
Efficiency improvements in coal-fired generation in the Reference Scenario
Mostly supercritical in OECD, some in China
45%
Supercritical in OECD, some USC and IGCC; more supercritical in non-OECD countries
USC and IGCC in OECD; some USC and IGCC in non-OECD countries
OECD Non-OECD World average
40%
35%
30% 2006
2010
2015
2020
2025
2030
Note: Average efficiency of installed coal-fired capacity. USC refers to ultrasupercritical steam conditions. IGCC refers to integrated gasification combined cycle.
Trends in gas-fired generation World gas-fired generation reached 3 807 TWh in 2006, 20% of total electricity generation. While it is expected to continue to expand in the future, rising to 4 723 TWh in 2015 and to 6 404 TWh in 2030, growth will be constrained by high gas prices (see Chapter 1). Most of the growth is projected to occur in non-OECD countries, with the biggest increase in the Middle East. Gas-fired generation is expected to continue to grow in the OECD, although much more slowly than over the past decade. Although high natural gas prices are expected to constrain demand for new gas-fired generation, it still has advantages that make it attractive to investors, notably lower capital costs and a shorter construction time than most other technologies. Moreover, the operating flexibility of these plants makes them a very valuable component of the portfolio of a power company, while the usually high correlation between power and gas prices in liberalised markets reduces the extent of gas and power price risks.
4. The need to combine variable renewables with gas also has implications for CO2 emissions.
146
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Gas-fired power plants are expected to be used increasingly for mid-merit order and peak load, replacing oil to some extent. This implies a need for a large increase in gas storage capacity. The projected large increase in power from variable renewable energy sources will require flexible back-up capacity: CCGTs and OCTs are expected to be used for this purpose, but high gas prices will make back-up power more expensive.4 Gas turbines are also expected to be used in decentralised electricity generation. Fuel cells using hydrogen from reformed natural gas are expected to emerge as a new source of distributed power after 2020, producing less than 1% of total electricity output in 2030.
Trends in oil-fired generation About 6% of total electricity generation worldwide is based on oil products. This share has been declining slowly, due to high oil prices. In the OECD, the share of oil in total electricity generation was 4% in 2006; but it is much higher in a number of countries, notably in Mexico (22%), Greece (16%), Italy (15%), Japan (11%), Portugal (11%), Ireland (10%), Spain (8%) and South Korea (6%). In most other countries, it is less than 2%. The main oil product used for power generation in the OECD is fuel oil, which accounts for 64% of total oil-fired generation. In 2006, oil-fired generation in the OECD dropped sharply, by about 20%, in response to rising oil prices. In the long run, oil-fired generation in the OECD is projected to fall further, accounting for 1% of total generation by 2030. Use of heavy oil products is expected to fall particularly sharply, the main oil product still in use in 2030 being diesel oil, used for peaking and plant start up and in isolated areas. In non-OECD countries, oil-based electricity generation remains flat in absolute terms, representing a fall in oil’s share in total generation. As in the OECD, diesel oil is expected to become the main oil product. Oil-producing countries, notably in the Middle East, are projected to continue to reduce their reliance on oil, so that more can be available for export.
6
Trends in nuclear power World nuclear capacity reached 368 GW in 2006. This rose to 372 GW in 2007. About 309 GW of this capacity was in OECD countries. Between 2000 and 2007, about 31 GW of new capacity were added worldwide.5 In the OECD, new reactors started commercial operation in the Czech Republic, France, Japan, Slovakia and South Korea. Outside the OECD, new reactors were brought on line in Brazil, China, India, Pakistan, Romania, Russia and Ukraine. A total of 8 GW was shut down during the same period: in Bulgaria, Lithuania, Slovakia and Ukraine, following agreements with the European Union (4 GW); in the United Kingdom (16 reactors totalling 2 GW), Spain (the 141-MW Jose Cabrera reactor) and Russia (a 5-MW reactor) as plants reached the end of their operating lives; in Japan, the prototype 148-MW Fugen reactor, because of high operating costs; and in Sweden (a second reactor at the Barseback plant, following one that was shut down in 1999) and Germany (two reactors with a total capacity of 980 MW) in implementation of policies to phase out nuclear power.6
A total of 31 GW of nuclear capacity is now under construction in different parts of the world (Table 6.2). The construction of some of these reactors started a number of years ago and was subsequently halted, mainly for economic reasons. In Argentina, the 5. This increase in capacity includes uprates (through upgrades of existing reactors). 6. Data are taken from the IAEA’s PRIS database, available at www.iaea.org/programmes/a2/.
Chapter 6 - Power sector outlook
147
© OECD/IEA, 2008
Over the past few years, a large number of countries have expressed renewed interest in building nuclear power plants, driven by concerns over energy security, surging fossil-fuel prices and rising CO2 emissions. Few governments, however, have taken concrete steps to build new reactors, beyond those that have had active nuclear power construction programmes in place for a long time.
government recently decided to complete the construction of a second reactor at the Atucha nuclear power plant following energy shortages; construction originally started in 1981 but had been put on hold. In Bulgaria, the government now plans to complete two reactors at the Belene plant, the construction of which had originally started in 1987. In Russia, work has been restarted on three of the seven reactors under construction in the 1980s. In Ukraine, there are plans to complete two unfinished reactors, construction of which started in the mid 1980s. In the United States, the Tennessee Valley Authority plans to complete a second reactor at the Watts Bar plant, on which construction started in 1972. In Iran, one reactor at the Bushehr power plant, which had been abandoned in the 1970s, is now almost complete. In total, these reactors account for a third of the total nuclear capacity under construction.
Table 6.2
Nuclear power plants under construction as of end-August 2008
Country
Number of reactors
Argentina
1
Capacity (MW) 692
Bulgaria
2
1 906
China
6
5 220
Chinese Taipei
2
2 600
Finland
1
1 600
France
1
1 600
India
6
2 910
Iran
1
915
Japan
2
2 166
Korea
3
2 880
Pakistan
1
300
Russia
7
4 724
Ukraine
2
1 900
United States
1
1 165
Total
36
30 578
Most of the nuclear capacity now under construction is in the Asia-Pacific region. China has the largest amount of capacity under construction in the world. India, Japan and South Korea are planning significant increases in their nuclear power capacity. World nuclear capacity is projected to rise to 397 GW by 2015 and to 433 GW by 2030. Installed capacity in 2030 is 18 GW higher than projected in WEO-2007 because the nuclear gain from assumed higher fossil-fuel prices offsets the effect of lower global demand for electricity. Most of the increase is in Asia, where China, India, Japan and South Korea are projected to nearly double their installed nuclear capacity, from 75 GW to 142 GW. Prospects for nuclear power have also picked up in the United States. Since 2002, a number of policies have been put in place to address investor concerns and some 17 companies and consortia are actively pursuing licenses for more 148
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Note: Installed capacity is net (electricity only). Source: IAEA PRIS Database (available at www.iaea.org)
than 30 nuclear power plants. In the Reference Scenario, installed nuclear capacity in the United States increases from 100 GW to 115 GW. In the European Union, nuclear capacity drops from 131 GW in 2006 to 89 GW in 2030, mainly as a result of phase-out policies in Germany, Belgium and Sweden.
Trends in renewable energy Higher fossil-fuel prices and increasing concerns over energy security and climate change are expected to encourage the development of renewable energy for electricity production in many parts of the world, particularly where incentives are already in place. Electricity generated from renewable energy sources worldwide amounted to 3 470 TWh in 2006, or 18% of total output. In the Reference Scenario, it rises to 4 970 TWh in 2015 and to over 7 700 TWh in 2030, 23% of total electricity production. Trends in renewable energy are discussed in more detail in Chapter 7.
6
Trends in CO2 capture and storage Carbon capture and storage (CCS) is a promising technology for carbon abatement, even though it has not yet been applied to large-scale power generation. Widespread deployment of CCS depends on developments in legal and regulatory frameworks, financing mechanisms, international co-operation, technological advances and public awareness. Recommendations on how to accelerate the deployment of CCS were developed in 2007 under a joint International Energy Agency-Carbon Sequestration Leadership Forum (IEA-CSLF) initiative, involving wide consultation with interested parties. These recommendations include launching 20 full-scale CCS projects by 2010 and making CCS eligible for funds from the Clean Development Mechanism (IEA-CSLF, 2007). Market mechanisms alone will not be sufficient to achieve the demonstration programme on the scale required. Another challenge is financing the necessary CO2 transport infrastructure. Recent legal and regulatory developments include amendments to the London Protocol and the OSPAR (Oslo-Paris) Convention to allow for storage of CO2 in formations below the international waters that are covered by those treaties. The European Commission’s climate change and renewable energy package, released in January 2008, includes a new directive on the EU Emissions Trading Scheme (ETS), which addresses CCS among other matters. It proposes new legislation to encourage CO2 storage. Also, under the European Union’s Zero Emissions Technology Platform, the FLAGSHIP programme will provide guidance on criteria for selecting 10 to 12 demonstration projects covering all the technology blocks of CCS (power plant and capture, and CO2 transport and storage) for evaluation by the European Commission.7
7. The Zero Emissions Technology Platform is a joint initiative between European industry, scientists and NGOs with support from the European Union to accelerate the development of low-carbon power-generation technologies and, in particular, CCS.
Chapter 6 - Power sector outlook
149
© OECD/IEA, 2008
In 2008, the StatoilHydro Snohvit project started injecting CO2 into a 2 600-metre deep formation, becoming the world’s fourth large-scale CCS project. In November 2007, the United Kingdom announced a national competition for government co-funding of a
post-combustion CCS project, covering as much as 100% of the additional capital and operating costs involved. Several projects have been announced, varying in scale from industrial prototypes to 1 200 MW scale, with target dates between 2010 and 2017 (Table 6.3).
Recent CCS proposals and developments
Company/project
Fuel
Technology
Output
BP-Rio Tinto DF4, Abu Dhabi
Gas
NGCC + EOR + Storage in saline aquifer
420 MW
2013
China Huaneng Group (CHNG), GreenGen, China
Coal
IGCC + shift + precombustion
100 MW
2015
E.ON Killingholme, Lincolnshire, UK
Coal
IGCC+shift+precombustion (may be capture ready)
450 MW
Ferrybridge, Scottish and Southern Energy, UK
Coal
PC (supercritical retrofit) + post-combustion capture, retrofit GBP 250 m, capture GBP 100 m
500 MW
E.ON Kingsnorth Kent, UK
Coal
Supercritical retrofit
FutureGen, United States
Coal
IGCC + shift + precombustion
275 MW
$1.5 bn
2012-2017 (under restructuring)
Hypogen, EU
Coal
Pre-combustion + H2
Approx. 250 MW
€1.3 bn
2014-2016
Coal
IGCC + shift + precombustion
1 000 MW
GE / Polish utility
Cost
£1 bn
Start date
2011 2011
2 × 800 MW £1 bn
2012
Karstø, Norway
Natural gas
NGCC + post-combustion 384 MW amine, potential storage in the oilfield – EOR
2012 (capture)
Mongstad, Norway
Natural gas
CHP with post-combustion CCS 280 MW Phase 1: Pilot capture 350 MWh Phase 2: Full-scale transport + storage
Phase 1: 2010 Phase 2: 2014
Nuon, Eemshaven, Netherlands
Coal/biomass /natural gas
IGCC with option to capture
1 200 MW
After 2011 (decision in 2009)
Coal
IGCC + shift + precombustion
Approx. 900 MW
2010
Powerfuel, Hatfield Colliery, UK VattenFall, Jänschwalde, Germany
Lignite
Stanwell, Queensland, Australia
Coal
New oxyfuel boiler, existing 2 × 250 MW boiler to be retrofitted with post-combustion capture
2015
IGCC + shift + precombustion, storage in saline reservoir
2012
100 MW
Notes: Shift is an essential reaction in IGCC technology using CO2 capture, involving a catalyst that converts carbon monoxide to carbon dioxide through a reaction with water. Capture ready refers to whether an existing plant can be readily converted to a CCS plant. This includes a number of conditions, such as space availability for additional capture equipment and proximity to a CO2 storage site. Amine solvents are used in post-combustion capture. Source: IEA (2008b).
150
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Table 6.3
In the Reference Scenario, CCS makes only a minor contribution. Stronger policies will need to be put in place — especially a widely applicable carbon constraint — if a significant amount of CCS capacity is to be commissioned before 2030 (see Part C).
New capacity and investment needs in infrastructure Total installed power-generation capacity worldwide in the Reference Scenario is projected to rise from 4 344 GW in 2006 to 5 697 GW in 2015 and to 7 484 GW in 2030. Total gross capacity additions amount to 4 528 GW over the period, with 1 691 GW installed by 2015.8 This is an average of 190 GW of new capacity per year through to 2030 — about the same as over the past few years. Cumulative global power-sector investment over 2007-2030, including generation, transmission and distribution, is $13.6 trillion (in year-2007 dollars), just over half of all energy-sector investment. Some $6.8 trillion of investment is required in generation, while transmission and distribution networks together need $6.8 trillion, of which twothirds goes to distribution. The largest investment requirements, exceeding $3 trillion, arise in China. Investment needs are also very large in OECD North America and Europe. Table 6.4 summarises the capacity additions and investment needs by region over the Outlook period. Table 6.4
6
Projected capacity additions and investment needs in power infrastructure in the Reference Scenario Investment, 2007-2015 ($2007, billion)
Investment, 2016-2030 ($2007, billion)
Capacity Power Transmission Distribution Capacity Power Transmission Distribution additions generation additions generation (GW)
OECD
514
982
278
656
1 107
2 467
403
922
North America
215
379
121
260
480
1 136
238
512
Europe
221
457
93
281
465
1 048
94
286
Pacific
78
146
65
115
163
283
71
124
Non-OECD
1 177
1 215
589
1 285
1 730
2 177
837
1 793
E. Europe/ Eurasia
137
180
55
183
159
274
51
173
Asia
781
794
433
894
1 170
1 379
596
1 231
574
521
296
612
718
753
299
618
Middle East
78
59
32
67
160
135
71
146
Africa
59
59
28
58
91
159
47
97
121
123
41
84
149
230
72
148
1 691
2 197
867
1 941
2 837
4 644
1 239
2 716
China
Latin America World
8. As Figure 6.4 shows, 613 GW of new capacity are now under construction.
Chapter 6 - Power sector outlook
151
© OECD/IEA, 2008
(GW)
Electricity generating costs Trends in construction costs There have recently been sharp increases in construction costs of new power plants, particularly in OECD countries. For non-OECD countries, fewer cost data are available, making it difficult to draw firm conclusions on the extent of the increases experienced there. The sharpest increases have been in the United States, where the construction cost of a new supercritical coal plant has doubled over a few years. Companies planning to build nuclear power plants there have announced construction costs at least 50% higher than previously expected. Similar trends, although somewhat less pronounced, are evident in other OECD countries. Table 6.5 provides an overview of these trends as seen by a power plant manufacturer. The main causes of higher costs are as follows: Increases in the cost of materials: The prices of metals typically used in power plants (iron, steel, aluminium, copper) have substantially increased since 2003/2004 (Figure 6.7) and the prices of some special steels used in power plant manufacturing have increased even faster than the average steel price shown in the chart. Cement prices are also reported to have gone up. These material costs have increased primarily because of high global demand for commodities and manufactured goods, higher production and transportation costs (in part owing to high fuel prices), and a weakening US dollar (The Brattle Group, 2007).
Index (2000 = 100)
Figure 6.7
Commodity price indices
500
Aluminium Copper
400
Crude oil Iron ore
300
Steel
200 100 0 2000
2001
2002
2003
2004
2005
2006
2007
2008
Increase in energy costs: High energy prices affect the manufacturing and transportation cost of power plant equipment and components. Increase in demand: As their reserve margins are shrinking, OECD countries are entering a new investment cycle. Outside the OECD, strong growth in electricity demand is pushing up orders for new plant. Tight manufacturing capacity: Reportedly, manufacturers of power plants are often not able to fulfil orders quickly because of a lack of capacity and a shortage of 152
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Sources: IMF (www.imf.org); CRU (cruonline.crugroup.com); and IEA databases.
skilled engineers. Steam boilers, gas turbines, wind turbines and photovoltaic panels are all in short supply. Many manufacturers claim that their order books are full for the next three to five years. At present, the nuclear industry faces bottlenecks in the construction of large steel forgings (an essential component of reactors). Increases in labour costs: A shortage of engineering, procurement and construction contractors in some regions and rising labour costs, particularly in non-OECD countries, is pushing up total project costs. Table 6.5
Factors influencing power plant pricing Price trend since 2003
6
Civil engineering Construction materials (cement, steel, etc.) Labour costs (local, international) Power plant Main mechanical components (boiler, generator, turbine) Other mechanical equipment (e.g. piping) Electrical assembly and wiring: – cables – transformers
(approx. + 270%) (+150%) (+150%) (+60%-90%)
Engineering and plant start up Additional factors Transportation and logistic costs Power plant demand Cost of capital and inflation Note: A vertical arrow denotes large increases, a slanted arrow moderate increases, and a flat arrow no increase. Source: Siemens AG.
Cost and efficiency assumptions
9. The cost and efficiency assumptions by technology used in this WEO are available at www.worldenergyoutlook.org.
Chapter 6 - Power sector outlook
153
© OECD/IEA, 2008
The power-plant construction-cost assumptions used in this Outlook have been substantially revised upwards (following a revision in WEO-2006) to take account of recent trends and industry expectations.9 Because the uncertainties surrounding future construction costs are so large, our newly revised costs for most types of power plants are generally assumed to apply from the onset and remain constant through to 2030. There are, however, some qualifications to this. Some cost reductions are assumed for new technologies, notably for IGCC and renewables. The construction costs of coal-fired power plants in China now are several times lower than in OECD countries. They are assumed to increase over time, reflecting increases in labour costs and more stringent environmental requirements, though they remain substantially lower than in the OECD even in 2030. China’s emergence as a major supplier of coal-fired
power plants to the rest of Asia and markets further afield (such as Russia and Africa) competing with well-established manufacturers in international markets, could drive down construction costs for all regions. Fossil-fuel prices have soared in recent years and are assumed to remain high. Higher plant construction and fuel-input costs result in much higher levelised generation costs over the projection period than in past WEOs. The electricity produced in new plants expected to come on line in the next few years will cost much more in OECD countries than was projected in WEO-2006, driving up end-use electricity prices. Generating costs in non-OECD countries are also much higher, but how much final prices are affected will depend on what subsidy-reform policies their governments adopt. As illustrated in Figure 6.8, by comparison with coal, nuclear or wind generation, gasfired generation is by far the most expensive in 2015 and 2030 in the United States, the European Union, China and India on the assumption of continuity of present policies.10
Dollars (year-2007) per MWh
Figure 6.8
Electricity generating costs in selected regions
200
Coal Gas
160
Nuclear Wind
120 80 40 0 2015 2030 United States
2015 2030 European Union
2015 2030 China
2015 2030 India
Note: Costs include a carbon value of $30 per tonne of CO2 in the European Union. In 2015, coal refers to supercritical steam. In 2030, coal refers to IGCC for the United States, ultrasupercritical steam for Europe and China, and supercritical steam for India. Gas refers to CCGT. Source: IEA analysis.
10. Wholesale power prices in European and North American markets are already in the same order of magnitude as the costs calculated here. 11. A value of $30 per tonne CO2 is assumed for ETS over the Outlook period. This is equal to €22 per tonne CO2, assuming an exchange rate of 0.73 euros per dollar (average value in 2007).
154
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
In the European Union, an assumed carbon value of $30 per tonne of CO2 adds around $30 per MWh to the generating cost of coal-fired plant and around $15 per MWh to the cost of power from CCGT plants.11 As a result, the difference in total cost between the two types of generation is small. Coal is the cheapest way to generate electricity
in the absence of a carbon value. In the United States, coal is accordingly the cheapest way to produce electricity, followed by wind power. Nuclear power in the United States is more expensive than coal and wind, but expectations about carbon constraints and a large need for firm base-load power are expected to support the development of new nuclear power stations. In China, coal is by far the cheapest way to produce electricity because of low construction costs and cheap fuel. Coal is also the cheapest way to generate electricity in India (as in many other non-OECD countries), though it is more expensive than in China.12 Generally, gas-fired generation becomes even more expensive over time. Coal-based generating costs fall slightly between 2015 and 2030 because coal prices are assumed to go down after 2015.
6 S P O T L I G H T
What are the risks of investing in the power sector? Projected generating costs (Figure 6.8) are but one criterion used by investors to decide on whether to invest and on the type of technology. Power companies increasingly use portfolio investment-valuation methodologies to take into account risks over their entire plant portfolio, rather than focussing on the technology with the lowest stand-alone projected generating cost. Investors may accept different risk profiles for different technologies, depending on project-specific circumstances. Concern about both climate change and structural change in the industry have introduced new risks for electricity operators. In relation to climate change, the risk is of uncertainty as to future policy and its implementation. The structural risk is more fundamental, since it involves the transfer of risk from the consumer (or tax payer) to the investor. In the traditional, vertically integrated public service model, the supply company was often a monopoly and could count on recovering the investment and the target return (in practice, governments often imposed price controls, for political reasons).
12. Coal transport distances in China and India can significantly increase fuel costs, adding to generating costs.
Chapter 6 - Power sector outlook
155
© OECD/IEA, 2008
In the competitive situation now existing in most OECD countries and several non-OECD countries, risks have, to some extent, moved from rate and tax payers to competing market players. This perception of increased risk drives up the investor’s required rate of return, a factor influencing the cost assumptions now embedded in the calculations from which our projections derive.
Electricity prices Electricity prices paid by final customers have increased in real terms in many OECD countries since the beginning of this decade, after falling during the 1990s (Figure 6.9). Sharply higher fuel-input costs, which make up a significant share of the total cost of power supplied to end users, are the principal reason, especially since 2003. The impact of fuel price increases on electricity prices is most direct for gas, which accounts for the largest part of the total costs of generating electricity. Moreover, electricity prices are set by marginal generation costs and — in more and more markets and on more and more occasions — CCGTs are the marginal generation source. Coal and uranium costs have also increased markedly, though in the case of uranium the cost of the fuel accounts for only about 5% of the total cost of generation.
Index (1990 = 100)
Figure 6.9
Indices of real end-use fuel and electricity prices in OECD
160
Gas Oil
140
Coal Electricity
120 100 80 60 1990
1992
1994
1996
1998
2000
2002
2004
2006 2007
Source: IEA databases.
Over the projection period, electricity prices are expected to increase, reflecting fuel price increases and higher construction costs. On average, OECD electricity prices are projected to be 11% higher in 2030 than in 2006. 156
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
In the European Union, the Emissions Trading Scheme (ETS) has also contributed to higher electricity prices (Figure 6.10). The price of carbon on the ETS fell sharply in 2006 as it became clear that CO2 allowances were plentiful for the first phase of the scheme, which ran until end-2007. This price drop was reflected in a marked fall in wholesale electricity prices in several European markets, illustrating the emerging tight interaction between the ETS and wholesale electricity trade. Wholesale prices rebounded in 2007 and 2008 with higher prices for carbon under the second phase of the scheme.
European wholesale prices of electricity and EU emission allowances
140
70
120
60
100
50
80
40
60
30
40
20
20
10
0 Jul/05
Jan/06
Jul/06
Jan/07
Jul/07
Dollars per tonne of CO2
Dollars per MWh
Figure 6.10
ETS* - 2007 (right axis) ETS* - 2008 (right axis) Electricity - United Kingdom Electricity - Nordic Electricity - Germany
6
0 Jan/08 Apr/08
* ETS — European Trading Scheme.
Chapter 6 - Power sector outlook
157
© OECD/IEA, 2008
Sources: EEX (www.eex.com/en/); Elexon (www.elexon.co.uk/); Nord Pool (www.nordpool.com/asa/); and PowerNext (www.powernext.fr/).
© OECD/IEA, 2008
CHAPTER 7
RENEWABLE ENERGY OUTLOOK H
I
G
H
L
I
G
H
T
S
Renewable energy sources are set to expand rapidly. Excluding traditional biomass use, their share of global primary energy demand is projected to climb from 7% in 2006 to 10% by 2030 in the Reference Scenario. This results from lower costs as renewable technologies mature, assumed higher fossil-fuel prices, which make renewables relatively more competitive, and strong policy support. The renewables industry has the opportunity to exploit this development to eliminate its reliance on subsidies and to bring emerging technologies into the mainstream. World renewables-based electricity generation — mostly hydro and wind power — is projected to more than double over the Outlook period, its share of total electricity output rising from 18% in 2006 to 23% in 2030. Renewables overtake gas to become the second-largest source of electricity, behind coal, before 2015. In the OECD, the total projected increase in renewable electricity generation is bigger than the combined increase in fossil fuel-based and nuclear power generation. The share of renewables in electricity generation increases by ten percentage points to 26% over the Outlook period. Global output of wind power is projected to increase eleven-fold, becoming the second-largest source of renewable electricity after hydro by 2010. The largest increase is in the European Union, where the share of wind power reaches 14% in 2030, accounting for 60% of the increase in total EU electricity generation between 2006 and 2030. Biomass, geothermal and solar thermal provided around 6% of total global heating demand in 2006. This share is projected to increase to 7% in 2030. Where resources are abundant and conventional energy sources expensive, renewablesbased heating can be very cost competitive with conventional fossil-fuel heating systems.
Total cumulative investment in renewable energy supply in 2007-2030 amounts to $5.5 trillion (in year 2007 dollars). The greater part of this investment is for electricity generation. Renewables account for just under half of the total projected investment in electricity generation.
Chapter 7 - Renewable energy outlook
159
© OECD/IEA, 2008
The share of biofuels in the total supply of road transport fuels worldwide is projected to rise from 1.5% in 2006 to 5% in 2030, spurred by subsidies and high oil prices. Most of the growth comes from the United States, Europe, China and Brazil. Second-generation biofuels are expected to become commercially viable but still make only a small contribution to total biofuels supply towards the end of the projection period.
Global trends in the use of renewable energy The use of renewable energy sources has been expanding rapidly in recent years and this trend is set to continue over the projection period. Investment in renewable energy sources for electricity, heating and in biofuels has increased considerably. However, if traditional biomass is excluded,1 renewable energy still accounts for a small share of the energy mix. In 2006 it met just 7% of global primary energy needs. The share was around 6% for heat demand (mostly from the direct combustion of biomass) and 1% for transport (biofuels derived from biomass). Renewables accounted for 18% of total electricity generation, the majority of it coming from hydropower plants (16% of total electricity generation). In the Reference Scenario, the share of renewables (including modern biomass) in the global primary energy mix increases from 7% in 2006 to around 10% in 2030. In the electricity sector, the generating costs of coal- and gas-fired power plants remain high, because of both high fuel prices and high construction costs.2 Over the projection period, oil, gas and coal prices are assumed to remain high by historical standards (see Chapter 1). While the construction costs of some renewables have also increased recently, the costs of renewable energy systems are expected to fall in the coming decades, mainly as a result of advances in technology and declining capital costs as their use expands. Demand for renewables for heating and cooling, and demand for biofuels are also expected to be boosted by high fossil-fuel prices. There is a good opportunity during the next decade or so for the renewables industry to prove it can become cost competitive without relying on subsidies and for the technologies to enter the mainstream. A carbon price on fossil fuels would make renewables even more competitive (see Part C). Despite the improved environment for emerging renewables sources and technologies, many barriers to their deployment remain. These could hold back long-term growth of the sector. They include: the relatively high costs of some technologies in the absence of subsidies; relatively limited research and development until recently; growing concerns about the impact on food availability of the use of crops for energy; a lack of skilled labour and policymaking capacity; regulations that discourage variable and distributed powergenerating systems; inadequate investment in networks; and scepticism on the part of some major incumbent players in the energy sector about the viability of renewables.
1. The use of fuelwood, dung and agricultural residues in very inefficient stoves or on open fires for cooking and heating. 2. See Chapter 6 for a discussion of recent increases in power plant construction costs.
160
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Biomass (including traditional biomass) is expected to remain by far the single most important primary source of renewable energy for decades to come. In the Reference Scenario, total world demand for biomass rises from 1 186 Mtoe in 2006 to about 1 660 Mtoe by 2030, although its share of world primary energy demand falls slightly from 10.1% in 2006 to 9.8% in 2030. Biomass differs from other renewable energy resources in that it can be a substitute for all fossil-fuel based products, using a wide range of technologies to convert the range of resources into heat, electricity and liquid fuels (Figure 7.1). It can be used directly in traditional ways for heating and cooking, or indirectly using modern conversion technologies. An estimated 60% of current biomass use is in the form of traditional biomass such as scavenged fuelwood, dried animal dung and agricultural residues used on open fires and in crude, low-efficiency stoves to provide basic cooking and heating services.
The quality of data on biomass use is often very poor, making future projections uncertain. Major uncertainties in projecting future demand for biomass for energy purposes include competition for land use (Box 7.1), the rate of improvement in crop yields, water availability for crop production, the effects of climate change and the development of advanced conversion technologies. Most of the biomass consumed in 2030 still comes from agricultural and forest residues, but a growing share comes from purpose-grown energy crops — mainly for making biofuels. A growing share is also projected to fuel combined heat and power (CHP) plants. Contribution of biomass to world primary energy demand, 2006
Figure 7.1
Global biomass consumption 1 186 Mtoe
7 Traditional biomass for direct heating and cooking 724 Mtoe
Modern biomass 462 Mtoe
Biofuels 24.4 Mtoe (40 Gl ethanol, 3 Gl biodiesel)
On-site heat 293 Mtoe
Industry 188.6 Mtoe
Electricity and district heat 80.7 Mtoe (239 TWh)
Losses 63.9 Mtoe
Buildings 104.4 Mtoe
Note: Electricity and district heat and industry include combined heat and power.
Land use required to grow biomass for energy
Of the 13.2 billion hectares (Gha) of the world total land area, more than 10% (1.5 Gha) are currently used to produce arable crops and over a quarter (3.5 Gha) are in pasture for meat, milk and wool production from grazing animals. Annually, around 7 to 8 Mha of forest land is converted to agriculture. Crops grown specifically for biofuels currently utilise around 1% of total agricultural land (FAO, 2008; WWI, 2007). This 1% share is the same in Brazil, where over 40% of the total gasoline demand is provided by ethanol based on sugarcane for feedstock and more is exported. At 38%, however, the land in Brazil, of which only a quarter is currently cultivated, is much higher than the world average of 10%. Other countries in Latin America and Africa also have the potential to expand their agricultural production with the aid of modern farm-management techniques and improved crop varieties, without resorting to deforestation. Shortage of water, either as natural rainfall or irrigation, can be a major constraint on increased energy crop production in some regions. Globally, some of the 1 Gha of marginal and degraded lands unsuitable for food production (due to rising salinity levels, for example, in Australia) could be reclaimed for productive use by growing selected energy crops.
Chapter 7 - Renewable energy outlook
161
© OECD/IEA, 2008
Box 7.1
Renewables for electricity Renewables-based electricity generation is expected to grow substantially over the coming decades, benefiting from high fossil-fuel prices, declining investment costs and government support. In the Reference Scenario, world renewable electricity generation is projected to increase from 3 470 TWh in 2006 to 4 970 TWh in 2015 and to 7 705 TWh in 2030. Before 2015, electricity from renewable energy sources is projected to overtake gas as the world’s second-largest source of electricity behind coal, rising from 18% in 2006 to 20% in 2015 and 23% in 2030 (Figure 7.2). The increase is far more substantial in OECD countries, where the share of renewables in electricity generation increases by ten percentage points, to 26%, over the Outlook period.
World
Figure 7.2
Share of total electricity generation from renewables by region in the Reference Scenario Hydro
2006
Other
2015 2030
OECD
2006 2015
Non-OECD
2030 2006 2015 2030 0%
5%
10%
15%
20%
25%
30%
Note: Other includes biomass, wind, solar, tide and wave power.
In the period to 2015, most of the increase in renewables generation is expected to come from hydro and onshore wind power. Both continue to grow between 2015 and 2030. Biomass and offshore wind power also grow briskly during this period. Solar power (both photovoltaics and concentrating solar) see their share of electricity output grow in many countries. In the OECD, the total projected increase in renewable electricity generation is bigger than the combined increase in fossil fuel-based and nuclear power generation (Figure 7.4). 162
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Hydropower remains the leading renewable source of electricity worldwide (Figure 7.3). But despite the projected large increase for hydropower in absolute terms — it nearly doubles in non-OECD between 2006 and 2030 — its share of total electricity generation falls over the projection period, even in non-OECD countries. The use of other renewables in electricity generation in non-OECD countries increases faster than hydropower. The share of all renewables in electricity generation in non-OECD countries increases from 21% in 2006 to 22% in 2030.
TWh
Figure 7.3
Increase in world electricity generation from renewables in the Reference Scenario
1 200
Hydro
1 000
Wind onshore Wind offshore
800
Biomass
600
Geothermal Solar PV
400
Concentrating solar
200
Tide and wave
7
0 2006-2015
TWh
Figure 7.4
2015-2030
Increase in OECD electricity generation by energy source in the Reference Scenario
1 200
Geothermal, tide and wave
1 000
Solar Biomass
800
Wind 600
Hydro
400
Nuclear Fossil fuels
200 0 2015-2030
The costs of power generation from renewables are expected to fall over time (Figure 7.5). The main reason is increased deployment, which accelerates technological progress and increases economies of scale in manufacturing the associated equipment, resulting in lower investment costs (Figure 7.6). The costs of the more mature technologies, including geothermal and onshore wind power, are assumed to fall the least. The investment costs of hydropower remain broadly unchanged. In reality, there is a wide range of generation costs for each technology, based mainly on the assumed lifetime, proximity to demand and local resource availability. Wind Chapter 7 - Renewable energy outlook
163
© OECD/IEA, 2008
2006-2015
power onshore, for example, can be generated for $30 to $40 per MWh in New Zealand, where many good sites near to cities enjoy a high mean annual wind speed, above 10 m/s. This enables wind to compete with hydro and gas without subsidies. Typical sites in Denmark and Germany, however, have mean wind speeds closer to 7 m/s, so that the same wind turbine generates only around half the electricity output per year that it would if built in New Zealand. The resulting relatively high generating costs require supporting policies to enable wind power to compete. In addition, since the generation costs of renewable electricity, other than from biomass, are largely dependent on the investment cost, the choice of discount rate applied adds to the wide cost range. The different characteristics of the output from different renewable energy technologies must also be taken into account when comparing renewable generating costs and when comparing them with fossil fuels and nuclear.3 The service provided by large hydro, notably when storage is used, makes it competitive with traditional peak-load and mid-load plants, which have higher generating costs than base-load plants. This increases the value of hydropower. By contrast, supply from wind and solar power depends on the availability of wind and sun. The output from these resources is variable, but if it correlates with peak load (for example, solar energy is available at peak hours in California, Japan and Southern Europe) the value of the electricity produced is higher.
Projected generating costs of renewable energy technologies in the Reference Scenario
Dollars (year-2007) per MWh
Figure 7.5 700 600 500 400 300 200 100
Biomass
Concentrating solar
2030
2015
2006
2030
2015
2006
2030
2015
2006
2030
2015
2006
2030
2015
2006
2030
2015
2006
Geothermal Wind onshore Wind offshore
Solar PV
3. See Figure 6.8 in Chapter 6 for the generating costs of coal-fired, gas-fired and nuclear power plants.
164
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Hydro
2030
2015
2006
0
Dollars (year-2007) per kW
Figure 7.6
Projected investment costs of renewable energy technologies in the Reference Scenario
8 000 7 000 6 000 5 000 4 000 3 000 2 000
Hydro
Geothermal Wind onshore Wind offshore
Biomass
Concentrating solar
7
2030
2015
2006
2030
2015
2006
2030
2015
2006
2030
2015
2006
2030
2015
2006
2030
2015
2006
2030
2015
2006
1 000
Solar PV
Note: Offshore costs increase with greater depth and distance from the shore.
Hydropower After two decades of stagnation, hydropower is making a comeback. Some 170 GW of hydropower capacity is under construction worldwide implying a significant increase of generation over the next decade (Figure 7.7). Most of this capacity is being built in non-OECD Asian countries such as China, India and Vietnam. The remaining potential in non-OECD countries is still large. Several non-OECD countries are re-focusing on this domestic source of electricity, driven by rapidly expanding power demand, the high cost of fossil fuels, growing concerns about pollution and greenhouse-gas emissions, and the perceived benefits to supply security by diversifying the electricity-generation mix. In Latin America, however, the amount of capacity under construction falls far short of that sought by government, notably in Brazil. Only 5 GW are being built, whereas the government expects installed capacity of 76 GW to double by 2030 (MME, 2007). We project hydropower capacity in Brazil to reach 104 GW in 2030 in the Reference Scenario.
Hydropower generation is projected to increase from 3 035 TWh in 2006 to over 3 730 TWh by 2015 and 4 810 TWh by 2030. Nevertheless, the share of hydropower in total electricity generation continues its downward trend, falling from 16% now to 15% in 2015 and 14.5% in 2030 as other power-generation technologies grow at faster rates. Chapter 7 - Renewable energy outlook
165
© OECD/IEA, 2008
Interest and support for hydropower projects in non-OECD countries are growing among international lenders and the private sector. In the OECD, many of the best sites have already been exploited and environmental regulations severely constrain new development. Most of the projected increase there comes from large-scale projects in Turkey and Canada. Several other OECD countries are providing incentives for developing small and mini-hydropower projects.
Figure 7.7
Hydropower capacity under construction by region 3%
3% 4% China
6%
Other Asia India
7%
Latin America
Total = 167 GW
OECD
9% 55%
Africa Russia
13%
Rest of world
Note: Includes power plants considered as under construction in 2007. Source: Platt’s World Electric Power Plants Database, January 2008 version.
Wind power Wind power has been growing at around 25% per year over the past few years. Worldwide, installed wind power capacity rose from about 6 GW in 1996 to 74 GW in 2006 and to 94 GW in 2007, with over 60% installed in OECD Europe. Germany has the largest installed capacity in the world, reaching 20.6 GW in 2006 and over 22 GW in 2007. Germany and Spain together account for about two-thirds of installed wind power capacity in OECD Europe. The rapid growth of wind power in these two countries is largely due to feed-in tariff systems,4 which offer high buy-back rates and guarantee a market for the output from wind farms. Globally, wind power accounts for a very small share of total electricity generation (around 0.7%), but this share is much higher in some countries (Table 7.1). Table 7.1
Top ten wind-power markets, 2006
Country
Installed capacity (GW)
Wind power production (TWh)
Share of total generation (%)
Germany
20.6
30.7
4.9
Spain
11.6
23.0
7.7
United States
11.6
26.7
0.6
India
6.3
8.0
1.1
Denmark China
3.1 2.6
6.1 3.9
13.4 0.1
Italy
2.1
3.0
1.0
United Kingdom
2.0
4.2
1.1
Portugal
1.7
2.9
6.0
France
1.6
2.2
0.4
4. A feed-in tariff is the price per unit of electricity that a power company has to pay for the purchase of renewable electricity which it is obliged by law to take from private generators. The tariff rate is fixed by the government, usually at a level high enough to encourage investment in electricity generation from renewables.
166
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Sources: IEA databases; GWEC (2007).
The United States has recently emerged as a major growth area for wind power and became the world’s second-largest market in 2007. Although incentives have been in place for a long time (a federal production tax credit [PTC] was introduced in 1992), very little growth in wind power was seen until recently. Instability of policy led to the PTC expiring, being extended, then later being renewed several times, heightening uncertainty and risks for investors. The current PTC is again due to expire at the end of 2008, after a one-year extension. The projections in this Outlook assume that the PTC will continue to be renewed, as it has been in the past. More recently, state-level incentives and obligations, and rising fossil-fuel prices have encouraged growth in wind power. Most of the growth has been in Texas and California, but wind power is also expected to grow in other states in the future. In the past, many projects in the United States faced siting difficulties, which resulted in them either not obtaining a licence or in very high costs, which made the developers abandon the project. Uncertainties remain about how fast the overall system and transmission network will be upgraded to accommodate large volumes of wind power. At the moment, not enough investment is going into the grid, which is impairing reliability. We assume that some of these difficulties will be overcome in the near future, facilitating the development of wind power.
7
Outside the OECD, India and China are important growing markets for wind power. India now has the fourth-largest installed capacity in the world and the largest outside the OECD. Wind power is also growing fast in China, where capacity has increased considerably since 2005. Both countries boast flourishing wind-turbine manufacturing industries. Global wind power output is projected to increase from 130 TWh in 2006 to more than 660 TWh in 2015 and 1 490 TWh in 2030. This increase will make wind the second-largest source of renewable electricity, after hydro, by 2010. Its share in total electricity generation rises from less than 1% in 2006 to 2.7% in 2015 and 4.5% in 2030. The share varies significantly between regions. In the European Union, it reaches 14% in 2030, accounting for 60% of the increase in total electricity generation between 2006 and 2030. In the United States, the share of wind rises ten-fold to 6.2% by 2030, meeting almost 30% of incremental demand for electricity between 2006 and 2030.
Grid integration of wind power and other variable renewables The rising share of wind power in electricity generation is posing a challenge for power companies in integrating this capacity into the network where penetration levels Chapter 7 - Renewable energy outlook
167
© OECD/IEA, 2008
Offshore turbines account for only a very small share of total wind-power capacity today, but this share is expected to increase substantially by 2030 as costs decline. Globally, offshore wind power production reaches about 350 TWh in 2030, up from only 3 TWh at present. The most substantial increase occurs in the European Union, where 27% of wind power could come from offshore installations by 2030.
are high. Wind power, like electrical power from wave, ocean currents, run-of-river hydro, solar photovoltaics and concentrating solar power systems, is different from conventional generating capacity because of its variability. These power outputs fluctuate according to the prevailing conditions of weather, season and time of day. There are, therefore, additional costs associated with the integration of these technologies into the power-supply system, including those associated with necessary back-up capacity and operation,5 and grid access (IEA 2008a).6 These extra costs can add up to $13 per MWh to the generating cost of wind power. Potential problems due to variability affecting the reliability of an electricitysupply system are manageable when the shares of wind and other variable renewables-based power capacity are small, but they become more pronounced as the shares rise. Technical problems can be largely overcome if the system is sufficiently flexible in responding quickly to large fluctuations in both supply and/ or demand. This can be the case if the system includes a balanced portfolio of power-generation technologies and fuels, substantial hydropower capacity with dam or pumped storage, arrangements for load-shedding to offset both peak demands and periods of lower power supply from variable renewables, good interconnection between neighbouring systems to reduce balancing requirements, and energy storage facilities (Box 7.2). Nonetheless, in a liberalised power market, higher shares of variable renewables, perhaps above 10% to 15% of total generation, may need some form of government regulation which mandates the integration of a specified share of renewables into the system.
Solar photovoltaics Worldwide, electricity generation from solar photovoltaics (PV) is still tiny, though the potential is very large. Installed global photovoltaics capacity was around 6 GW in 2006, with most of it installed in Germany (2.9 GW), Japan (1.7 GW) and the United States (0.6 GW). Electricity generation from PV integrated into buildings is, of course, more economically attractive in areas with abundant sunshine and high electricity prices. PV power is particularly valuable when maximum PV generation — around the middle of a sunny day — coincides with peak electricity demand (such as for air-conditioning). PV is already a cost-effective option in remote off-grid applications, either as a stand-alone system or as a part of a mini-grid.
5. Alternative generating capacity must be available to meet demand when there is no wind. Due to the current difficulty of accurately predicting wind patterns more than 36 hours in advance as required by some transmission operators, the provision of a reliable supply of power within a system can be complex and costly. Where gate-closure times are only 1 to 2 hours, the problem is largely overcome. 6. Wind turbines can be connected to the local grid at low-voltage levels, a practice that may save grid losses but adds to the complexity of system control and operation.
168
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
PV technology has the highest investment cost of all commercially deployed renewable energy sources. Investment costs for a system (including solar collection modules and the other components of the plant), currently range from around $5 500 to
$9 000 per kW, depending on the type of solar cell, scale of installation and local labour costs. In spite of big falls in module costs in recent years and increases in the efficiency of commercial cells, PV generation remains relatively costly. Further technological advances and ultimately achieving economies of scale will depend on government subsidies. However, new semi-automated PV manufacturing plants, with high annual production capacities of around 300 MW, are bringing down module costs per unit. In the Reference Scenario, the investment cost of a PV system is projected to fall to $2 600 per kW by 2030. There is optimism in the photovoltaics industry that very large cost reductions are in prospect, well before 2030.
Box 7.2
Energy storage
7 Where variable-output renewable energy systems, such as wind and solar, account for a major part of power generation within a system, storing surplus energy to meet later demand is one way to reduce the need for back-up generating capacity: storage avoids the need for power to be generated instantly to meet demand. Currently, storage is rarely the cheapest way of dealing with variability, but it could become increasingly viable as the price of carbon rises. Electricity cannot be stored directly (except in small capacitors), so it needs to be converted into other forms of storable energy, such as chemical energy in batteries, the potential (gravitational) energy of water in pumped-hydro storage systems, or mechanical energy as compressed air or in flywheels. Advanced systems are under development, such as ultra-capacitors, superconducting magnetic systems and vanadium redox batteries, but they tend to have low energy densities and poor durability, and add high costs to the system.
Producing hydrogen as an energy carrier from renewable electricity systems — by electrolysis, thermo-chemically from biomass, or by other means — can be another form of storage. This can be at a large scale in aquifers or caverns, or on a small scale involving compressing and storing the gas in cylinders or transmission pipes. The hydrogen can be used to generate electricity in fuel cells on demand but the overall efficiency of the system tends to be low, at around 40%. As a result of this and the high current investment costs for fuel cells, the unit cost is high.
Chapter 7 - Renewable energy outlook
169
© OECD/IEA, 2008
The capacities of storage systems, the cost of their construction and the overall costs per unit of energy stored vary widely. The capacity of facilities for pumped hydropower and compressed air in underground cavities can reach hundreds of MW. The re-conversion efficiency of pumped storage can be higher than 70%. Flywheels can be more efficient, but generally only at a very small scale. Batteries used either individually or in banks are around 70% efficient with capacities of just a few MW at the most. They tend to be more costly than the other options, at around $20 per MWh. The other key variable of a storage technology is the energy discharge time, which can range from seconds to hours.
Electricity generation from photovoltaics in the Reference Scenario is projected to reach 42 TWh in 2015 and 245 TWh in 2030, up from just a few TWh in 2006. Most of the increase comes from decentralised systems integrated into buildings where the generation cost competes against the retail electricity price and not the lower wholesale market price. Generous feed-in tariffs and other government incentives, together with falling PV costs and rising grid electricity prices, drive this growth.
Concentrating solar thermal power Several concentrating solar power (CSP) options are currently available for largescale electricity generation, though they are at varying stages of development. The parabolic-trough is already commercially available, while the parabolic-dish and central-receiver technologies are still at the demonstration stage. These could eventually achieve higher conversion efficiencies and lower capital costs than the parabolic-trough technology. Today’s oldest commercial parabolic-trough projects were developed in the United States in the early 1980s. CSP installation has been very slow since then, due to the high capital cost of the system. However, following a few parabolic-dish and central-receiver demonstration projects in Australia, Greece, Israel, Spain and Switzerland, as well as in the United States, interest in CSP has started to rise again. Total installed capacity of CSP was 354 MW in 2006. In 2007, a new 64 MW plant came on line in the United States and a 50 MW plant in Spain. Another 255 MW is under construction in Italy and Spain. Several new projects are being planned in other countries, including China, Iran, Jordan and Malta, with over 2 GW of total installed capacity likely by 2010. CSP generation is projected to reach 11 TWh by 2015 and 107 TWh by 2030. Current power-generating costs under the best conditions are around $150 to $250 per MWh or even less where the sun shines persistently. The lower end of the generation cost range is close to that of gas-fired generation at current gas prices, but CSP is generally more expensive than coal-fired generation (in the absence of a carbon value), nuclear and wind power. The economics are expected to improve in the future, falling to around $100 per MWh in sunny areas by 2030.
Electricity production from geothermal energy reached 60 TWh in 2006. Although, at the global level, geothermal power production has a very small share of the mix, it meets a significant share of the total electricity demand in Iceland (26%), El Salvador (20%), Philippines (18%), Costa Rica and Kenya (14% each), and Nicaragua (11%). As more fields are exploited, technologies are developed for deeper drilling into high-temperature fields and to enhance geothermal systems (hot dry rocks), power generation is projected to triple to almost 180 TWh by 2030, doubling geothermal’s share of 0.3% of the global electricity-generation mix in 2006. Most of the increase is expected to come from projects in the United States and non-OECD Asia. 170
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Geothermal power
Tide and wave power Ocean-energy power systems currently make the smallest contribution of all renewable energy sources. Tidal power stations are in operation in France, Canada, China, Russia and Norway. Wave power and ocean-current technologies are at an early stage of commercialisation, but there is strong interest on the part of several governments to develop them further. The United Kingdom is the leading country in which a specific market-based energy policy has been developed to encourage ocean energy (IEA-OES, 2007). This policy includes capital grants, purchase obligations, tradable certificates, guaranteed prices, and regulatory and administrative rules. In the Reference Scenario, tidal and wave power is projected to reach 14 TWh worldwide in 2030, compared with less than 1 TWh now.
7
Biomass for electricity Among all renewable energy sources, biomass is the largest contributor to global primary energy supply, amounting to 1 186 Mtoe in 2006. Of this, around 7% was used to produce 239 TWh of electricity. Approximately 44% of this electricity was generated in electricity-only generation plants, at around 20% conversion efficiency. The remainder was generated in CHP plants, where the heat is either used on-site (as in pulp and paper manufacturing plants) or sold and transported off-site, often for use in district-heating schemes in buildings and by industry. In the Reference Scenario, biomass use in power generation and heat plants increases by over 5% each year to reach close to 290 Mtoe of primary energy in 2030. Allowing for a projected increase in fuel-to-electricity conversion efficiency over time (typically 18% to 22% for woody biomass electricity-only plants in 2006, rising to 25% by 2030), this will require approximately 1 200 Mt of biomass (approximately 40 million truck loads) to be delivered annually to the generating plants. Over 860 TWh of electricity is projected to be generated by 2030, 60% of it in OECD countries. The share of biomass in global power generation doubles to 2.6%. The biggest increases in generation occur in the United States, Europe and China.
Biofuels
7. Preliminary data for 2007 shows biofuels demand at just over 0.8 mb/d (adjusted for energy content).
Chapter 7 - Renewable energy outlook
171
© OECD/IEA, 2008
Final demand for biofuels (transport fuels derived from biomass) worldwide reached 24.4 Mtoe (0.6 mb/d adjusted for energy content) in 2006, up from only 10.3 Mtoe in 2000 and 6 Mtoe in 1990.7 Biofuels represented only around 1.5% of total roadtransport fuel demand in 2006. North America is the world’s largest biofuels consumer (Table 7.2), followed by Latin America and the European Union, and most of the recent growth in biofuels use has come from the United States. Demand there grew by some 23% per year on average between 2000 and 2006. The United States overtook Brazil as the world’s largest biofuels consumer in 2004.
Table 7.2
Final consumption of biofuels by region in the Reference Scenario (Mtoe) 2006 16.9 11.3 5.5 0.1 7.5 0.0 0.8 0.0 0.0 6.6 24.4 5.5
OECD North America Europe Pacific Non-OECD E. Europe/Eurasia Asia Middle East Africa Latin America World European Union
2015 49.5 32.9 15.8 0.8 24.0 1.1 7.6 0.3 0.7 14.4 73.5 16.6
2030 72.5 46.8 24.7 0.9 46.0 1.5 17.9 0.8 1.1 24.7 118.5 25.9
2006-2030* 6.3% 6.1% 6.5% 10.3% 7.9% 16.8% 14.0% n.a. n.a. 5.6% 6.8% 6.7%
* Annual average rate of change.
In the Reference Scenario, world biofuels supply is projected to climb to 118 Mtoe in 2030 (3.2 mb/d), meeting 5% of total road-transport fuel demand8 (Figure 7.8). The projections for biofuels have been revised upwards substantially since WEO2007, due to new policies in several countries and because of higher oil prices than assumed last year. We assume in the Reference Scenario that the biofuel mandates in China and the European Union will be met after a lag of a few years but that biofuels in the United States in 2030 will attain only about 40% of the very ambitious target in the 2007 Energy Independence and Security Act. Asia and OECD Europe experience faster rates of growth, but in absolute terms these increases trail those World biofuels consumption by type in the Reference Scenario
120
6%
100
5%
80
4%
60
3%
40
2%
20
1%
0 2006
2015
2030
Biodiesel Ethanol Share of biofuels in road transport fuel consumption (right axis)
0%
8. The biofuels demand projections in this Outlook are based on a detailed region-by-region analysis of policy mandates, inter-fuel competition and supply potential. More detail about the prospects for supply to 2013 can be found in the IEA’s Medium-Term Oil Market Report, released in July 2008 (IEA, 2008b).
172
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Mtoe
Figure 7.8
in the larger North American market. Biofuels demand in the OECD Pacific region remains modest. Growth in Latin America is moderate, a consequence of the sizeable share of the market in Brazil already held by biofuels. In 2030, about 28% of total Brazilian road-transport fuel demand is projected to be met by biofuels (up from 13% today) and some 8% of the market in the United States (2% today). Ethanol currently accounts for a much larger share of the global biofuels market than biodiesel. Over the Outlook period, demand for biodiesel is set to grow faster than ethanol, so that the share in energy terms of total biofuels coming from ethanol falls from 83% in 2006 to 79% in 2030 (Figure 7.8). The United States and Brazil remain the largest ethanol consumers. The European Union and Asia see the fastest growth in demand for biodiesel. Second-generation biofuels, based on ligno-cellulosic feedstocks using enzyme hydrolysis or biomass-to-liquids gasification technologies, are expected to become commercially viable, but only make a minor contribution in the second half of the projection period. Biofuels are almost entirely used for road transport, but recent demonstration projects in the United Kingdom, Australia and Japan indicate there may be potential for using them on a large scale in aviation and shipping.
7
Several countries have aggressive policies in place for encouraging the production and use of biofuels, including the United States, where the Energy Independence and Security Act 2007 mandates a significant increase in both first- and secondgeneration biofuels use by 2020. China has a target to 2020 and the European Union has a target for biofuels to meet 10% of road transport demand by 2020. Australia, New Zealand, Colombia, South Africa, Thailand, Japan, Indonesia, Mexico and Canada also have mandates for ethanol blends. Some countries have recently started to reconsider their ambitious biofuels policies, including imported biofuels. This is due to concern about the environmental sustainability of production, overall greenhouse-gas emissions based on life-cycle analysis, and the impact on land use and food prices (see Spotlight below). The European Union is reviewing its policies, while India and Indonesia have already scaled back their incentives for biofuels. S P O T L I G H T
Soaring global food prices have recently sparked riots in many countries, including Haiti, Côte d’Ivoire, Ethiopia, Madagascar, Philippines and Indonesia. Food prices accelerated sharply in 2008, with grain prices more than doubling since January 2006. Over 60% of the rise in food prices has occurred since January 2008. Competition from biofuels production is one of many factors contributing to higher food prices. Other factors include: increasing dietary demand for meat and milk products, with consequent increased demand for animal fodder; high energy prices, which have pushed up the cost of fertilizer and other farm expenses, as well as food processing and distribution costs; poor harvests, due to extreme weather events such as droughts and destructive storms;
Chapter 7 - Renewable energy outlook
173
© OECD/IEA, 2008
To what extent are biofuels driving up the price of food?
the declining value of the US dollar; and international and national agricultural policies. The introduction of foodexport restrictions by some nations has restricted global supply and aggravated food shortages. The recent increase in biofuels production has certainly boosted the demand for specific crops and, in some countries, has displaced land that could have been used for growing food. However two of the commodities most often associated with today’s food crisis, wheat and rice, are not major sources of biofuel feedstocks. On the other hand, higher prices for crude oil, gas and electricity have greatly affected the cost of producing, processing and transporting food worldwide. The US Government Accountability Office concludes that food aid organisations spend 65% of their budgets on transportation. As a result of the rise in crude oil prices, the cost per food mile worldwide has soared. While it seems clear that rising biofuels output is contributing to the increase in food prices, there is no consensus on how large this factor is. A US Council of Economic Advisors study found the contribution of biofuels to the price increases of agricultural commodities to be only 3% (CEA, 2008) whereas a recent OECD report holds biofuels production responsible for 15% of the food price increases (OECD, 2008). A study published by the World Bank concludes that recent increases in biofuels production (and related consequences of low grain stocks, large land-use shifts, speculative activity and export bans) explain 70% to 75% of the increase in food prices (Mitchell, 2008). It is very difficult to compare these studies because of different methodologies, different time periods and other factors. More analysis is needed to determine the true impact of biofuels production, because of the very many assumptions and uncertainties involved in assessing its effects on food prices, commodity stocks, oil prices, and direct and indirect land-use change over various time horizons.
Today, global trade in ethanol represents just under 20% of total ethanol demand, but this share has been steadily rising from about 12% in 2002 (F.O. Lichts, 2008). Brazilian exports, including volumes re-exported from countries in the Caribbean Basin Initiative, account for about 45% of global trade. Brazil is the world’s largest exporter of biofuels; the United States is the world’s largest importer. The Netherlands, Germany and the United Kingdom are the largest importers in the European Union. Biodiesel derived from 174
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Agricultural commodities will see much less competition between food, fibre and biofuels supply if productivity continues to rise at the growth rate seen in the past five decades as a result of better farm management, new technologies and improved crop varieties. Since so much available arable land needs irrigation, increased biofuel production in some regions will have implications for water resources. The uneven global distribution of natural resources results in regional differences, with many countries experiencing major land and water shortages. These differences impact on which countries have a comparative advantage in biofuels production.
palm oil, exported from Indonesia and Malaysia to the European Union, accounts for the majority of biodiesel trade. The United States and Brazil also export soybean biodiesel to EU countries. Numerous life cycle analysis studies have been conducted, but the various benefits of biofuels over conventional oil-based fuels for transport are still uncertain. Methods of allocating the emissions between co-products (such as high-protein animal meal from oilseeds and distillers grain from corn) are being determined. The greenhouse-gas emissions reduction potential per kilometre travelled is variable. It is highest in Brazil for ethanol produced directly from sugarcane because the fossil-fuel needs for processing are lower, as the sugar is fermented directly and the crushed stalk of the plant (known as bagasse) is used to provide heat and power in the production process rather than fossil energy. For each unit of sugar-based ethanol produced, only about 12% of a comparable unit of fossil energy is required (IEA, 2004). Overall, CO2 emissions calculated on a lifecycle basis are relatively low, at about 10% to 20% those of conventional gasoline.
7
A recent IEA analysis of 60 studies indicated that the conversion of sugarbeet into ethanol in Europe can yield reductions in well-to-wheels emissions of typically between 40% and 60%, compared with gasoline (OECD, 2008). Estimates for the net reduction in greenhouse-gas emissions that are obtained from rapeseed-derived biodiesel in Europe, compared with conventional automotive diesel, are in a similar range. For corn-based ethanol in the United States, the emission savings are much lower, at only 10% to 30% per kilometre compared with petroleum-based fuels. However, many of these studies do not include emissions resulting from direct and indirect land-use change, which can be significant but are hard to measure.
There is no clear consensus about when second-generation technologies will become commercially competitive, even with high oil prices. The key factors in achieving deployment are to prove the optimum technologies at a commercial scale, increase the scale of production, exploit the learning curve, and apply process optimisation and integration techniques. In the United States, Sweden, Canada and elsewhere, demonstration ligno-cellulosic ethanol plants have recently been built. There are also plans to build several small-scale cellulosic bio-refineries over the next 10 to 15 years in the United States. These will produce liquid transportation fuels such as ethanol, as well as bio-based chemicals and other bio-products used in industrial applications. The large Borregaard bio-refinery plant in Norway, employing 900 staff and using spruce Chapter 7 - Renewable energy outlook
175
© OECD/IEA, 2008
The successful development of second-generation technologies using non-food biomass feedstocks would allow the use of a much wider array of feedstocks, including dedicated crops such as grasses and fast-growing trees. The reductions in greenhouse-gas and polluting emissions could be significantly higher than those associated with firstgeneration ethanol and some of the fuels produced may be more suited for aviation, marine and heavy transport applications. Ligno-cellulosic ethanol technologies by bio-chemical conversion using enzymes are the focus of a considerable amount of research into such second-generation biofuels, notably in the United States. Thermochemical conversion, by gasification and the Fischer-Tropsch synthesis of the gases into petroleum substitutes, is also under evaluation at a demonstration scale in Germany and elsewhere.
logs as feedstock, has been operating for several decades, producing fuel ethanol and other products. The US Department of Energy is providing $385 million between 2007 and 2011 to support the development of six demonstration plants with the aim of making cellulosic ethanol cost-competitive with gasoline by 2012. The capital investment necessary to build cellulosic ethanol facilities, however, is significantly more than that of grain- or sugarcane-based facilities. As a result, the full commercialisation of these technologies hinges on further major cost reductions (IEA, 2008c). In the Reference Scenario, second-generation biofuels are not expected to penetrate the market on a fully commercial scale before 2020.
Renewables for heat Renewable energy systems directly provided around 300 Mtoe of heat in 2006, excluding the heat supplied from the combustion of traditional biomass or from heating appliances using renewable electricity. Biomass, solar thermal and geothermal currently account for around 6% of total heat demand worldwide. Renewables heating is projected to grow to 516 Mtoe in 2030, boosting its share of total heat demand to about 7% by 2030. This increase comes mainly from increased use of biomass for space and process heating, and greater solar water heating in residential buildings. Commercial biomass used in boilers and stoves to provide space and process heat in the building and industry sectors amounted to 293 Mtoe in 2006 across all regions. This includes the heat from CHP installations that is consumed on site (mainly in industry). In the Reference Scenario, biomass consumption rises to 453 Mtoe in 2030. Most of the increase occurs in OECD countries, through growth in the number of CHP installations in industrial facilities and increased demand for the heating of buildings. In non-OECD countries, the use of modern biomass increases from 164 Mtoe in 2006 to 241 Mtoe in 2030, leading to the more efficient and sustainable use of the local resource of basic forms of biomass.
Geothermal heat, used directly, accounted for around 3 Mtoe of space heating in 2006. The United States, Sweden, China, Turkey and Iceland are the leaders in such direct geothermal heat use.9 In Iceland it contributes around 45% of the total end-use building 9. Geothermal heat use in China is not reported in IEA statistics, but it is estimated to be around 0.5 Mtoe.
176
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
There is considerable potential for solar water heating to grow in the OECD, since around 20% of total residential energy consumption is for heating water. Over 90% of OECD solar water heaters are installed in the residential sector, which is expected to remain the largest consumer to 2030, although use by the commercial sector will also grow. Those heaters already provide over 7.6 Mtoe of energy for water heating, with a small proportion used for swimming pool and space heating. There is also potential for increasing use of solar water-heating systems in non-OECD countries, since the heaters are often manufactured locally and cheaply. China alone has around 60% of the world’s total installed capacity and collected around 3 Mtoe of useful solar heat in 2006. Globally, solar collection used for water and space heating (not including the passive solar heating of buildings) is projected to rise to about 45 Mtoe by 2030, with widespread global distribution.
heat demand. A third of the energy from this source comes from deep bores and the rest from millions of domestic scale, shallow ground-source heat pumps. Sweden alone has around one third of a million geothermal systems installed and projections show a significant increase will occur worldwide. Direct use of geothermal heat is projected to reach 18 Mtoe in 2030. The costs of renewable heating technologies vary according to local conditions, particularly the availability of the resource and the size of the market. Where the resources are abundant and conventional energy sources are expensive, they can be very cost-competitive with conventional fossil-fuel or electric-heating systems and require few, if any, government incentives (IEA, 2007). In China, for example, they are a relatively cheap option, particularly in rural areas where gas and electricity are not widely available. There is a large potential to reduce the installed costs of most renewable heating technologies through technological progress, training, mass production (to exploit economies of scale) and improved operating performance. Government intervention can help address some of the barriers to wider deployment, including planning difficulties (see Part C).
7
Traditional biomass “Traditional biomass” refers to fuelwood, charcoal, agricultural residues and animal dung. Most of this type of energy is used for lighting and direct heat in poor countries in Africa, Asia and Latin America, especially in rural areas where access to affordable, modern energy services is limited.10 When used sustainably, the use of traditional biomass is not in itself a cause for concern. However the World Energy Outlook has for many years highlighted traditional biomass use because the way in which these resources are used to provide energy services — cooking, heating and lighting — often has a serious impact on health, economic productivity and the environment. More people die prematurely in poor countries from the health impacts associated with indoor air pollution from burning biomass than from malaria. In addition, a great deal of time and effort is devoted to fuel collection, usually by women and children. This restricts them from engaging in more productive activities, such as enrolling in school and running small enterprises. Exploitation of biomass resources can also result in local environmental damage, such as land degradation, deforestation and air pollution. These negative effects can be ameliorated with policies and programmes which promote improved cookstoves with chimneys, proper ventilation in kitchen design, efficient charcoal kilns and fuelwood plantations.
10. Chapter 15 provides an outlook for the use of modern fuels in oil- and gas-rich sub-Saharan Africa countries. See Chapter 8 in WEO-2007 for a discussion of household energy demand in China and Chapter 20 for India.
Chapter 7 - Renewable energy outlook
177
© OECD/IEA, 2008
Although accurate and detailed statistics are essential for proper policy and market analysis, very few governments report on the use of traditional biomass at the household level. In order to raise the awareness of household energy consumption in non-OECD countries, the IEA has attempted a global assessment and has estimated the number of people who rely on these fuels, based on national surveys. We calculate that 2.5 billion people in Africa, non-OECD Asia and Latin America rely
on the burning of biomass resources in inefficient devices to meet their cooking and heating needs. In the absence of new policies, this number of people will rise to 2.7 billion in 2030 (IEA, 2006). The use of biomass in Africa is mainly in the residential and services sectors for cooking and heating, and it is not expected to decline much over the Outlook period. Traditional use of biomass declines in non-OECD Asian countries, as rising incomes enable more households to switch to cleaner-burning fuels. Use of biomass in the residential and services sectors declines in Latin America.
Investment in renewable energy In the Reference Scenario, total cumulative investment in modern forms of renewable energy in the period 2007 to 2030 amounts to $5.5 trillion (in year-2007 dollars). Most of the investment in renewables — 60% of the total — is for electricity generation, followed by investment in renewables for heat and biofuels, which account for 36% and 4% of the total, respectively. Total investment in renewables for electricity generation in the Reference Scenario from 2007 to 2030 amounts to $3.3 trillion; this provides 1 617 GW of additional capacity, mainly from an increase in hydro and wind capacity. Average annual capacity additions for renewables-based generation in the period 2007 to 2015 are projected to be 59 GW: some 27 GW for hydro and 32 GW for other renewables. Total annual capacity additions amount to 188 GW over the same period. In 2016-2030, hydropower capacity additions are 24 GW per year and other renewables 49 GW per year, out of a total new generation capacity of 189 GW per year. Annual investment in renewables will climb accordingly. Over the whole of the projection period, the annual investment in new renewable electricity capacity is higher than that in fossil-fuel power plants (Figure 7.9). Renewables-based capacity accounts for 48% of total projected investment in electricity generation between 2007 and 2030.
Billion dollars (year-2007)
Figure 7.9
World investment in new power-generation plants by fuel in the Reference Scenario
400
Nuclear Fossil fuels
300
Renewables
200
0 2006
178
2015
2030
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
100
CHAPTER 8
ENERGY USE IN CITIES H
I
G
H
L
I
G
H
T
S
Cities are a dynamic and vital part of global culture and are the main engines of social, economic and technological development. According to UN projections, by 2030, cities will house 60% of the world’s population — equivalent to the total global population in 1986. The geographic distribution of urban population is set to change: while global urbanisation in the first half of the 20th century was dominated by European cities, the majority of urban residents today live in Asia, despite the relatively low proportion there of city residents. The scale and pattern of city energy use has significant implications both for energy security and global greenhouse-gas emissions. Alert to the climate-change challenge, some city authorities are actively involved in reducing energy use and CO2 emissions. About two-thirds of the world’s energy — an estimated 7 900 Mtoe in 2006 — is consumed in cities, even though only around half of the world’s population lives in urban areas. City residents consume more coal, gas and electricity than the global average, but less oil. Increases in urbanisation through to 2030 are projected to drive up city energy use to almost 12 400 Mtoe in the Reference Scenario. By 2030, cities are responsible for 73% of the world’s energy use. Some 81% of the projected increase in energy use in cities between 2006 and 2030 comes from non-OECD countries. Energy use per capita of city residents is slightly lower than the national average in the United States, the European Union and Australasia (Australia and New Zealand). By contrast, city residents in China use almost twice as much energy per capita as the national average, due to higher average incomes and better access to modern energy services.
By 2030, 87% of US energy will be consumed in cities, up from 80% in 2006. In the European Union that figure will rise from 69% to 75% over the Outlook period. Australasian cities’ share of energy consumption will rise from 78% to 80% by 2030 and Chinese cities account for 83% of Chinese energy consumption in 2030, up from 80% today.
Chapter 8 - Energy use in cities
179
© OECD/IEA, 2008
The gap between rural and city energy use per capita is assumed to narrow over the projection period in China, but increasing urbanisation pushes up substantially the share of China’s energy used in cities.
Why focus on cities? This chapter analyses the size and pattern of energy consumption in the world’s cities and towns today and projected developments through to 2030.1 Cities (including towns) currently use over two-thirds of the world’s energy and account for more than 70% of global CO2 emissions.2 Driven by growing urbanisation, city energy use is set to increase significantly through to 2030: in the Reference Scenario, energy use in cities rises to 73% of the global total and CO2 emissions to 76%. Given the scale of city energy use and CO2 emissions, if cities and their authorities were to take action to mitigate climate change, this growth could be curtailed. Cities are a dynamic and vital part of global culture and are the main engines of social, economic and technological development. From the small cities in the Pacific to the large mega-cities of Asia or North America, cities are home to an ever-increasing number of people. But to provide their populations with the myriad of services demanded, cities need large amounts of energy. Today, much of that energy is fossilfuel based. As the urban population, economic activity and wealth increase, urban energy use is set to grow. In 2008, half of the world’s population live in cities. According to UN projections, by 2030, cities will house 60% of the world’s population — equivalent to the total global population in 1986. At the same time, the geographic distribution of urban population is set to change: while global urbanisation in the first half of the 20th century was dominated by European cities, the majority of urban residents today live in Asia, despite the relatively low proportion there of city residents (Figure 8.1). Some of the fastest-growing cities are in Africa.
Urbanisation rate
Figure 8.1
Regional trends in urbanisation
100%
1980 2006
80%
2030
60% 40% 20% 0% OECD North Latin America America
OECD Pacific
OECD Europe
Middle East
E. Europe/ Eurasia
Asia
Africa
1. Throughout this chapter, “cities” refers to all urban areas, including towns. The terms “cities” and “urban” are used interchangeably. Box 8.1 describes the ways energy data and projections of energy use in cities were prepared. 2. See Chapter 16 for detailed CO2 emission projections for cities.
180
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Sources: UNPD, 2007a and 2007b.
Box 8.1
Methodological issues and city energy data
There is no international consensus on the definition of a city. For reasons of data availability, in this Outlook “city” refers to all urban areas, from “megacities” to smaller-scale urban settlements. City energy data are difficult to find, often incomplete and rarely in a format that allows comparisons between cities or with national data. Once data are acquired, analysis requires consideration of city boundary issues — what energy was consumed within the city, how to deal with energy supplied from outside the city (for example, power generation) and transit energy consumption (for example, vehicles passing through a city). Many cities are collecting their own data on energy consumption, but there is a lack of standard reporting methods. Various local government associations, for example the International Council for Local Environmental Initiatives (ICLEI), are attempting to establish more uniform data-collection methods. The global estimates of city energy use and CO2 emissions presented in this Outlook have been calculated on the basis of the detailed energy-consumption projections for the 21 WEO regions in the IEA’s World Energy Model. City energy use was calculated using detailed analysis from four regions selected on the basis of the importance of these regions and the availability of city energy data: the United States, the European Union (split into two), Australia and New Zealand combined, and China. These regions account for over half of the world’s current energy consumption. For each of these four regions a detailed analysis of 2006 data produced, for each fuel and by end-use sector, a per-capita ratio of city energy use to national energy use. These ratios were used to derive total final consumption in these four regions. The share of electricity consumed in cities was used to apportion national power generation, regardless of whether the electricity was generated in the city or not. This result was then used to calculate the primary energy used in power generation associated with the electricity consumed in cities.3 Each of the four modelling region’s ratios were calculated using methods appropriate to the data available.4 In addition to these four detailed regional analyses, selected major cities in other regions for which detailed data were available, such as Tokyo, Moscow and Delhi, were used to further refine the ratios. We applied the most appropriate per-capita ratios, and the UN urbanisation projections, to the energy projections of the 21 WEO regions to calculate energy demand in cities at the global level through to 2030. For OECD regions, the differential between rural and urban per-capita energy consumption is assumed to remain constant over the Outlook period. For non-OECD regions, we assume that the differential converges over the Outlook period towards the urban/rural differential in OECD regions. Energyrelated CO2 emissions were calculated using national carbon factors by sector and fuel (see Chapter 16 for detailed CO2 emissions projections).
4. A detailed definition of regions and a description of the methodology for modelling city energy use are available at www.worldenergyoutlook.org.
Chapter 8 - Energy use in cities
181
© OECD/IEA, 2008
3. Based on the assumption that the city power-generation mix and conversion efficiencies are the same as those at the regional level.
8
The scale and pattern of city energy use has significant implications both for energy security and global greenhouse-gas emissions. Alert to the climate-change challenge, some city authorities are actively involved in reducing energy use and CO2 emissions. For example, London has an aggressive climate-change policy involving a range of activities, such as working in partnership with private sector companies to design, finance, build, own and operate decentralised, low-energy and zero-carbon projects. The analysis of global city energy use presented here is, to our knowledge, the most in-depth ever attempted. In view of data deficiencies and methodological challenges, however, the results should be treated as no more than indicative.
Current and projected energy use in cities Our detailed analysis of current city energy use focuses particularly on four regions: the United States, the European Union, Australasia (defined here as Australia and New Zealand), and China. In the first three, city energy use per capita is roughly the same as the national average (Table 8.1). Cities in the European Union tend to use less energy per capita than US and Australasian cities, because of higher population densities, more extensive urban public transport systems, and more district heating. By contrast, cities in China are significantly more energy intensive per capita than the national average as a result of higher incomes and better availability of energy services compared with rural areas. The recent increase in Chinese national per-capita energy consumption — 8.5% per year between 2000 and 2006 — has been driven partly by urbanisation. During that period the urban population in China grew by 3.2% per year, adding a further 91 million inhabitants to Chinese cities. Table 8.1 Region
Overview of city energy use and urbanisation rate in regions and countries analysed in depth, 2006 Share of city primary energy demand in regional total
Ratio of city per-capita primary energy demand to regional average
Urbanisation rate
United States
80%
0.99
81%
European Union
69%
0.94
73%
Australasia
78%
0.88
88%
China
75%
1.82
41%
Based on the results of our analysis of primary energy use in cities in these four regions, we estimate that around 7 900 Mtoe was consumed globally in cities in 2006 (Table 8.2). This equates to 67% of the world’s primary energy demand in 2006, or more than four times that of China — the world’s second-largest energy consumer. The proportion of global energy consumed in cities is greater than the proportion of the world’s population living in cities. This is because city energy use is related to economic 182
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Note: Fuel breakdowns for each region are in the regional section.
activity, as well as city population. A large proportion of industrial and commercial activity, which consume a large proportion of total energy, is located in cities. For example, the majority of commercial buildings are located in cities. Urban energy use is also higher than rural energy use because, in the developing world (with urbanisation rates of 30% to 50%) residential per-capita energy use tends to be significantly higher in cities than in rural areas. This is because energy use is related to income and, as noted above in relation to China, city dwellers in the developing world tend to have higher incomes and better access to energy services. The income effect usually outweighs efficiency gains due to higher density settlement in cities. By contrast, city and rural residents in developed countries tend to enjoy similar levels of energy service access. Because of the rapid growth of cities in the developing world, their pattern of energy use will increasingly shape global energy use, though the rural/urban differential is expected to diminish somewhat.
Table 8.2
8
World energy demand in cities by fuel in the Reference Scenario 2006
2015
2030
Mtoe
Cities as a % of world
Mtoe
Cities as a % of world
Coal
2 330
76%
3 145
78%
Oil
2 519
63%
2 873
63%
Gas
1 984
82%
2 418
Nuclear
551
76%
Hydro
195
75%
Biomass and waste
280
24%
Other renewables
Mtoe
Cities as a % of world
20062030*
3 964
81%
2.2%
3 394
66%
1.2%
83%
3 176
87%
2.0%
630
77%
726
81%
1.2%
245
76%
330
79%
2.2%
358
26%
520
31%
2.6%
48
72%
115
73%
264
75%
7.4%
Total
7 908
67%
9 785
69%
12 374
73%
1.9%
Electricity
1 019
76%
1 367
77%
1 912
79%
2.7%
Globally, the share of natural gas (82%) and electricity (76%) consumed in cities is higher than the average of all fuels, and much higher than the share of the world’s population living in cities. This is due to the more extensive infrastructure in cities for energy distribution and higher appliance-ownership rates in developing cities relative to rural areas. Coal consumption in cities accounts for 76% of the global total, mainly from coal-fired power generation and coal use in urban-based industry in developing countries, such as China. The share of oil consumed in cities (63%) is smaller than the average of all fuels, due to higher penetration of electricity for heating and cooking, and wider use of urban public transport networks. The share of biomass and waste consumption, at 24% globally, is much lower in cities than in rural areas, consisting in OECD cities largely of biofuel consumption and biomass in industry and power generation. In developing countries, traditional biomass and waste is still used for Chapter 8 - Energy use in cities
183
© OECD/IEA, 2008
* Average annual growth rate.
domestic cooking and heating in cities, but consumption is much less per capita than in rural areas. Some 72% of other renewables are consumed in cities, mainly for power generation in OECD countries. In the Reference Scenario, global city energy use is projected to grow by 1.9% per year (compared to an overall global growth rate of 1.6% per year), to 12 374 Mtoe in 2030. Its share of global energy use rises from 67% to 73% (Figure 8.2). Non-OECD countries account for 81% of the growth in energy use in cities over the Outlook period.
Mtoe
Figure 8.2 18 000
World and city primary energy consumption in the Reference Scenario Non-OECD OECD
15 000 12 000 9 000 6 000 3 000 0 Cities
World 2006
Cities
World 2015
Cities
World 2030
United States Background Over the past century, the United States has evolved from a largely industrial economy, dominated by the large cities of the Northeast and Midwest, towards a more dispersed, services-oriented economy. This shift has been characterised by growth in suburban areas and smaller cities, along with the expansion of urban areas in the Southwest which are dependent on cars and air conditioning. The proportion of the US population living in urban areas has increased from 40% in 1900 to over 80% today, with most population growth over the 20th century occurring in urban areas (USGCRP, 2001). Over the next 20 years, urban population in the United States is expected to grow by approximately 1% per year, reaching 87% of the total population by 2030.
5. At market exchange rates.
184
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
The United States contributes just 5% of global population and 8% of urban population, but it is responsible for over 25% of global GDP,5 nearly 87% of which is generated in metropolitan areas.6 US energy policy is mainly controlled by the federal government,
though state-level authorities are increasingly influential. For example, although the United States has not ratified the Kyoto Protocol, several states have voluntarily set their own targets which meet or exceed Kyoto targets. California aims to reduce its emissions to 80% below 1990 levels by 2050. The Regional Greenhouse Gas Initiative in the eastern part of the United States and the Western Governors’ Initiative are multi-state climate change and greenhouse-gas emissions assessment and reduction initiatives. Three frequent drivers of local policy in the United States are the security of the local energy supply, the affordability of energy services, and the long-term economic prospects of the city. New York City’s 2004 Energy Policy Task Force report and 2007 Citywide Sustainability Plan were explicit in warning that the city would soon be unable to meet peak demand for power and in identifying what steps must be taken to remedy the problem. In other cases, such as in San Francisco, local environmental concerns have been pre-eminent, expressed particularly in efforts to force local utilities to reduce power-plant emissions levels. Several large cities, including Los Angeles, Austin and Seattle, actually own and operate the utility providing local gas and electric services, giving them significant influence over the types of fuels used to generate power.
8
Several mayors, including those of Seattle, New York, Salt Lake City, Chicago and San Francisco, have made climate and energy planning a central tenet of their administrations. They have also been vocal in rallying other cities to join in these efforts. In 2005, the mayor of Seattle spearheaded the formation of the US Conference of Mayors Climate Protection Agreement, an initiative whereby cities pledge to reduce local greenhouse-gas emissions to 7% below 1990 levels by 2012. As of July 2008, 850 mayors across the United States — representing 80 million people — have signed this pledge. In 2007, the mayor of New York hosted the second C40 Large Cities for Climate Protection conference, an international initiative involving 40 of the largest cities around the world. Later that year, the mayor of New York spoke at the Bali Climate Change Conference on behalf of mayors around the world, calling on international climate negotiators to reflect on the role cities can have in global climate-change mitigation efforts.
Reference Scenario projections
6. Defined as core urban centres and adjacent surrounding counties (US Census Bureau, 2000; BEA, 2007).
Chapter 8 - Energy use in cities
185
© OECD/IEA, 2008
City and urban areas account for 80% of total US primary energy demand (Table 8.6) driven by the high rate of urbanisation. This is significantly higher than the global average of around 67%. US cities tend to have slightly lower per-capita energy consumption than the national average. Most urban efficiency gains are attributable to the transportation sector, where each urban resident consumes 11% less transport energy than the average US resident.
Figure 8.3
US energy consumption in cities by state, 2006
State energy consumption – urban consumption % < 30% 61 - 70% 31 - 40% 71 - 80% 41 - 50% 81 - 90% 51 - 60% > 90% The boundaries and names shown and the designations used on maps included in this publication do not imply official endorsement or acceptance by the IEA.
The boundaries and names shown and the designations used on maps included in this publication do not imply official endorsement or acceptance by the IEA.
Source: IEA analysis.
186
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Per-capita energy consumption, urban: total 0.5 - 0.75 (urban state average) 1.26 - 1.41 (urban >> state average)
Table 8.3
US energy demand in cities by fuel in the Reference Scenario 2006
2015
2030
Mtoe
Cities as a % of national
Mtoe
Cities as a % of national
Mtoe
Cities as a % of national
20062030*
Coal
470
85%
512
88%
577
91%
0.9%
Oil
679
72%
668
75%
683
77%
0.0%
Gas
432
86%
454
89%
478
92%
0.4%
Nuclear
184
86%
199
89%
232
93%
1.0%
Hydro
22
86%
23
89%
24
93%
0.4%
Biomass and waste
68
85%
111
88%
170
92%
3.9%
Other renewables Total Electricity
11
85%
29
88%
65
91%
7.9%
1 865
80%
1 996
83%
2 229
87%
0.7%
277
86%
310
89%
376
93%
1.3%
* Average annual growth rate.
8
Total urban energy consumption in US cities is set to increase from 1 865 Mtoe in 2006 to 2 229 Mtoe by 2030, growing at 0.7% per annum, almost double the 0.4% per annum growth at the US national level. This growth is driven by projected growth of the urban population in the United States, from 81% today to 87% by 2030. However, improvements in energy intensity at the national level result in a drop in per-capita energy consumption of urban residents from 7.6 Mtoe today to 7.0 Mtoe by 2030. Findings for the United States as a whole mask strong regional variations in urban consumption, driven by large differences in individual city size, levels of urbanisation, climate, density, housing stock and per-capita income (Figure 8.3). In most coastal states, urban energy consumption is over 70% of the total; in many of the densely populated and highly urbanised states of the Northeast, urban energy consumption accounts for over 90% of state consumption. Per-capita consumption tends to be higher in urban areas east of the Mississippi and lower in urban areas west of the Mississippi.
European Union As one of the early centres of industrial transformation, the area that makes up the European Union is characterised by high levels of population density, urbanisation and energy use. With a current population of almost 500 million, accounting for 8% of the global population, the European Union generates 31% of global GDP.7 Population projections are flat throughout the Outlook period, while urbanisation is projected to continue growing from 73% today to 80% by 2030. Over the past decade, the European Union has been one of the driving forces behind international policy agreements for binding climate-protection targets. The European Union has set up an internal market for emission trading, which was operational before the Kyoto Protocol came in to force in February 2005. 7. At market exchange rates.
Chapter 8 - Energy use in cities
187
© OECD/IEA, 2008
Background
Compared with other world regions, the European Union has a distinctive form of multi-level governance uniting 27 member states. It does not have a truly common and comprehensive energy policy, though a growing body of EU legislation does affect the energy sector in member countries. European Union directives cover a range of issues, including co-generation using renewable energy (combined heat and power), internal markets in electricity and gas, and energy efficiency. Additionally, industrial point-source greenhouse-gas emissions are traded in the world’s largest multi-national and multi-sector market for emission trading, the European Union Emissions Trading Scheme (see Chapter 17), covering more than 10 000 installations of the energy and industrial sectors. Urban and regional climate-protection strategies operate in parallel with the efforts at the national and supra-national scale. Arrangements for, and the extent of, energy planning and policy at the urban and regional level vary considerably among member states, reflecting the heterogeneity of the structure of government and the levels of local participation in member states. Some members of the European Union have a federal organisation, with extensive powers exercised at the urban and regional level. Cities like Berlin, Bremen or Hamburg, for example, are federal states (Länder) in their own right, able to decide their individual energy programmes. Communes in the Nordic countries have considerable control over how to spend their tax income; and in the UK many planning decisions are devolved to the local authority or metropolitan level. European cities have given a lead, internationally, to efforts to mitigate climate change. Almost all EU capitals and large cities have adopted, or are currently discussing, local climate-change action plans and greenhouse-gas emissions reduction targets that are, at times, more ambitious than those at the European and national levels. A specific characteristic of the European institutional framework for energy and climate policy is the importance of city networks, many of them partly financed by the European Union, as a way of sharing ideas and best practice.
Reference Scenario projections
Total urban primary energy consumption in the European Union is projected to increase by 168 Mtoe over the projection period, reaching 1 427 Mtoe by 2030 — an average rate of growth of 0.5% per year. This is more than twice the rate of the European Union as a whole, which averages 0.2% per year and is accounted for by the growth in the urban share of the EU population to 80% by 2030 (Table 8.4). Urban energy consumption increases in absolute terms as the urban population rises, because improvements in energy intensity can not keep pace with the rise in population. Per-capita energy consumption increases from 3.5 Mtoe in 2006 to 3.6 Mtoe by 2030 (Figure 8.4). 188
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
In 2006, urban areas accounted for 1 259 Mtoe, about 69%, of the total primary energy demand of the European Union. Final energy demand on a per-capita basis was about 3.5 Mtoe in urban areas, slightly lower than the EU average of 3.7 Mtoe per capita. Average per-capita consumption in rural areas was even higher, at around 4.9 Mtoe.
The fuel mix in cities differs considerably from the EU average. A higher share of gridbased fuels, such as gas, heat and electricity, is consumed in cities, compared with the EU average for all fuels, whereas a lower proportion of coal, biomass and oil is used. Table 8.4
European Union energy demand in cities by fuel in the Reference Scenario 2006 Mtoe
2015
Cities as a % of regional
Mtoe
2030
Cities as a % of regional
Mtoe
Cities as a % of regional
20062030*
Coal
207
64%
214
66%
183
70%
-0.5%
Oil
397
59%
388
60%
384
64%
-0.1%
Gas
397
91%
456
92%
536
96%
1.3%
Nuclear
187
73%
171
74%
138
80%
-1.3%
Hydro
19
74%
25
76%
31
82%
1.9%
Biomass and waste
41
45%
61
46%
93
49%
3.4%
Other renewables
10
70%
29
72%
62
76%
8.1%
1 259
69%
1 344
71%
1 427
75%
0.5%
176
73%
204
75%
245
79%
1.4%
Total Electricity
8
* Average annual growth rate.
Mtoe
Figure 8.4
Per-capita energy demand in the European Union and EU cities in the Reference Scenario
3.9
EU average
3.8
Cities
3.7 3.6 3.5 3.4 3.3 3.2 2006
2015
2030
Australasia Both Australia and New Zealand have developed largely through agriculture and natural resource production, and are now characterised by a few cities that are large relative to the national population. Urbanisation rates have been traditionally high Chapter 8 - Energy use in cities
189
© OECD/IEA, 2008
Background
and continue to grow. In 2006, Australia had a population of 20.6 million people, with 13.1 million (63.5% of the total population) living in eight regional capital cities. In 2006, New Zealand’s population was 3.9 million people, with 1.9 million (48.5%) living in the three major urban areas. As most of the development of cities in Australia and New Zealand has taken place since the advent of the motor vehicle, the density of the residential urban population, approximately 1 500 people per square kilometre, is low. Cities tend to conform to a spread-out suburban model, with a denser inner ring (around a core that has become denser in the last decade) and gradual reduction of population density as the city expands. Most major Australian cities are governed by multiple local governments, each of which is responsible for a portion of the urban area. Local governments have the power to enact laws and implement policies relating to planning, transport, building, waste and other local issues relating to energy management and climate change. In general, local governments in Australia can implement policy that affects energy demand, both stationary and mobile, but not energy supply. Their role as the planning authority, with responsibility for approving residential and commercial developments, is their main source of influence in relation to energy consumption and CO2 emissions. New Zealand cities are governed by local governments, comprising of 12 regional councils, 74 territorial local authorities and four unitary authorities in total. Each local authority has some degree of control over energy-using activities. Typically, local governments (such as Christchurch and Auckland) have intervened on air quality issues and seek to take climate-change issues into consideration in developing other policies for which they are responsible. This has been particularly important in the development of the Long-Term Council Community Plans. Additionally, as the planning authority for wind farms and hydro developments, local governments have been at the forefront of the development of alternative energy systems. Local governments are not prevented from developing power generation facilities or operating an electricitysupply business but, to date, the focus of policy has been on managing demand through policy implementation or ambitious targets such as the zero net-emissions target of Wellington. The commercial and services sector together now dominate the Australian economy, accounting for 60% of GDP in 2006. This proportion of the economy has been increasing for a number of years (ABS, 2007). Most commercial activity is office-based and has relatively low energy intensity. Residential energy consumption, on the other hand, has been increasing steadily for many years, driven by increasing appliance-ownership rates. Moving people and goods around sparsely populated countries such as Australia and New Zealand requires a large amount of energy.
In 2006, the cities in Australia and New Zealand consumed 78% of total energy demand, totalling 109 Mtoe (Table 8.8). Cities consumed 80% of total electricity and almost all natural gas in the end-use sectors, as there is little local distribution of natural gas in rural areas. Energy use per capita in cities is lower than the national average, at 5.0 Mtoe in 2006, compared to 5.6 Mtoe for Australia and New Zealand as a whole. 190
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Reference Scenario projections
Table 8.5
Australasian energy demand in cities by fuel in the Reference Scenario 2006 Mtoe
2015
Cities as a % of regional
Mtoe
2030
Cities as a % of regional
Mtoe
Cities as a % of regional
20062030*
Coal
45
80%
49
82%
47
83%
0.2%
Oil
30
66%
34
67%
35
69%
0.7%
Gas
25
93%
28
95%
34
97%
1.3%
Nuclear
0
n.a.
0
n.a.
0
n.a.
n.a
Hydro
3
80%
3
82%
3
84%
0.8%
Biomass and waste
5
75%
7
77%
10
79%
3.5%
Other renewables Total Electricity
2
70%
3
67%
7
64%
5.9%
109
78%
124
79%
136
80%
0.9%
17
80%
20
82%
24
84%
1.4%
8
* Average annual growth rate.
The commercial sector is less energy intensive per capita than the residential sector, and industry and transport are the most energy-intensive sectors. Total urban energy demand in Australasia is projected to increase from 109 Mtoe today to 136 Mtoe by 2030, growing at 0.9% per year. This is slightly quicker than the 0.8% per year consumption growth for the region as a whole. In addition to an increased use of oil in the rural residential sector, there is an increase in the use of biomass, due to the proximity of fuelwood sources and lower population density reducing air pollution concerns.
China Background
In China, city governments have some influence over local energy supply and demand. one reason being that public utilities and many industries are still owned by city authorities. However, cities are firmly guided by national policies, which they are responsible for implementing. China’s 20% energy-intensity reduction target by 2010, in its 11th Five-Year Plan, illustrates the relationship between national and local government. This target was delegated to provincial level in the form of provincial performance targets. Although regional autonomy was strengthened through reforms in the 1990s, energy remains a strategic sector directed by the central government. Chapter 8 - Energy use in cities
191
© OECD/IEA, 2008
China contributes 20% of the global population and 17% of the global urban population. In 2006, China’s urban population was 545 million, more than the entire population of the European Union. The UN projections see the current rate of urbanisation in China — around 40% — reaching 60% by 2030. Given that Chinese cities are more energy intensive per capita than rural areas, this will have significant implications for the national energy system and global energy markets.
There is intense competition between cities in China to attract investment and, as a consequence, cities act to address concerns about energy security and local air pollution. Interventions by city authorities on energy matters tend to concentrate on four major fronts: Clean energy, especially by encouraging greater penetration of natural gas and limiting the use of coal in the household and commercial sectors. Action in the energy sector to facilitate economic structural change from primary to secondary industries and towards an enlarged tertiary sector. Promoting energy efficiency. Improving public and mass-transportation systems. Climate change is becoming an increasingly important policy priority for both the national government and city governments. A number of energy and environmental policies at the city level have an impact on climate issues. Examples include the local air-pollution control policy associated with the 2008 Beijing Olympics, Shanghai’s vehicle-supply control policy, and policies to increase investment in renewable energy. Some Chinese cities have also been involved in international efforts to develop low carbon cities, initiated by organisations such as the World Wildlife Fund and others. New cities are putting greater emphasis on energy and climate issues. Dontang Eco-City, which is being built at the mouth of the Yangtze River close to Shanghai, is one example. Differences in the percapita energy use and CO2 emissions of Chinese cities reflect a number of local conditions, of which scale and structure of the local economy are key factors (Figure 8.5).
Primary energy consumption (toe per capita)
Figure 8.5
Per-capita energy demand and gross regional product in selected Chinese cities, 2006
8 7
Hohhot
6
Yinchuan
Taiyuan
5
Dalian Shanghai
4
Guangzhou
3
Ningbo
Xining
2
Beijing
Fuzhou
1 Chongqing
0 0
2 000
4 000
6 000 8 000 10 000 12 000 Gross regional product (2006 dollars, MER) per capita
Source: Dhakal, 2008
The share of cities in China’s total primary energy demand in 2006 is estimated at 75% or about 1 424 Mtoe (Table 8.6). The per-capita energy use of urban areas in China is about 80% higher than in the country as a whole. On average, each urban citizen in 192
World Energy Outlook 2008 - GLOBAL ENERGY TRENDS TO 2030
© OECD/IEA, 2008
Reference Scenario projections
China consumes 2.6 Mtoe, compared with 1.4 Mtoe nationally, reflecting higher urban incomes. Coal provides about 87% of urban primary energy. Natural gas use in cities in China is increasing rapidly. Beijing leads this trend, and the share of natural gas has dramatically increased since 2000. Further urbanisation and economic development should lead to a convergence of urban and rural energy intensity. Cities will have an increasing role in shaping the energy profile of China, with urban areas providing great opportunities for large-scale policy interventions for managing the overall energy system. By 2030 cities in China are projected to consume 3 220 Mtoe. The proportionate effect on energy consumption of strong urbanisation growth is partially offset by the fact that increasing energy consumption in rural areas is also expected to grow strongly through the Outlook period, with the net result that the share of city energy use in China increases to 83% by 2030. Phasing out of coal, and greater use of natural gas and liquefied petrolum gas, in residential and commercial sectors in urban areas is set to increase. Table 8.6
2006
Coal
8
Chinese energy demand in cities by fuel in the Reference Scenario 2015
2030
Mtoe
Cities as a % of national
Mtoe
Cities as a % of national
Mtoe
Cities as a % of national
20062030*
1 059
87%
1 665
88%
2 206
90%
3.1%
Oil
271
77%
428
77%
648
80%
3.7%
Gas
40
81%
84
82%
158
84%
5.9%
Nuclear
12
84%
39
84%
67
87%
7.5%
Hydro
31
84%
52
84%
76
87%
3.8%
Biomass and waste
10
4%
12
5%
37
16%
5.6%
Other renewables
2
45%
9
62%
27
67%
12.2%
1 424
75%
2 289
79%
3 220
83%
3.5%
161
80%
314
80%
495
83%
4.8%
Total Electricity
Chapter 8 - Energy use in cities
193
© OECD/IEA, 2008
* Average annual growth rate.
© OECD/IEA, 2008
PART B OIL AND GAS PRODUCTION PROSPECTS
PREFACE
How much oil and gas is there in the ground and how much can be economically produced? Over the years to 2030, what investment will be required to exploit these reserves and can we be confident that it will be forthcoming? These are the questions addressed in Part B. The oil and gas is there, but the sums required to exploit them are daunting. Remaining reserves of oil continue to rise, despite growing consumption, and the gas-reserves position is still more comfortable. But the exploitation costs have risen fast and, as we become increasingly dependent on OPEC countries and their national oil companies, the opportunities for international oil companies diminish. If prices stay high, the depletion policies of some producing countries become increasingly restrictive. On the basis of a detailed field-by-field analysis of production trends at 800 of the world’s largest oilfields and the assessment of the prospects for reserves growth and new discoveries, Chapters 9, 10 and 11 examine in detail the evidence on available resources and the prospects for future supply. The demands are great. Between now and 2015 some 30 mb/d of new production capacity needs to be brought on stream to compensate for the decline in production from existing fields and meet the growth in demand. Chapter 12 similarly reviews the prospects for gas. Chapter 13 quantifies the investment needed to exploit oil and gas resources on the required scale and, together with Chapter 14, which looks at the structural changes taking place in the upstream industry, assesses whether the prevailing conditions warrant confidence that the investment will be forthcoming.
© OECD/IEA, 2008
Finally in this Part B, oil and gas prospects in oil- and gas-rich sub-Saharan Africa are analysed, exposing potential government earnings from oil and gas of some $4 trillion over the years to 2030 — many times more than would be required to bring basic modern energy services to the energy-poor of these countries. But will these government resources be devoted to this cause, which can underpin economic development?
© OECD/IEA, 2008
CHAPTER 9
TURNING OIL RESOURCES INTO RESERVES How much is left to produce? H
I
G
H
L
I
G
H
T
S
The world is far from running short of oil. Remaining proven reserves of oil and natural gas liquids worldwide at the end of 2007 amounted to about 1.2 to 1.3 trillion barrels (including about 0.2 trillion barrels of Canadian oil sands). These reserves have almost doubled since 1980. Though most of the increase has come from revisions made in the 1980s in OPEC countries rather than from new discoveries, modest increases have continued since 1990, despite rising consumption. The volume discovered has fallen well below the volume produced in the last two decades, though the volume of oil found on average each year since 2000 has exceeded the rate in the 1990s thanks to increased exploration activity (with higher oil prices) and improvements in technology. Ultimately recoverable conventional resources — a category that includes initial proven and probable reserves from discovered fields, reserves growth and economically recoverable oil that has yet to be found — amounts to 3.5 trillion barrels (leaving aside about 0.5 trillion additional barrels which might come from new sources not yet assessed and the application of new technologies). Only a third of this total has been produced up to now. Undiscovered resources account for about a third of the remaining recoverable oil, the largest volumes of which are thought to lie in the Middle East, Russia and the Caspian region.
Non-conventional oil resources are also large. Oil sands and extra-heavy oil resources in place worldwide amount to around 6 trillion barrels, of which between 1 and 2 trillion barrels may be ultimately recoverable economically. These resources are largely concentrated in Canada (mainly in Alberta) and Venezuela (in the Orinoco Belt). There is additional potential from oil shales, but their production cost and the environmental impact of their commercialisation are very uncertain.
Chapter 9 - Turning oil resources into reserves
197
© OECD/IEA, 2008
Future reserves growth will depend, to a large extent, on increases in the recovery factor, which is estimated to average about 35% worldwide today. Such increases, through secondary and enhanced oil recovery (EOR) techniques and other factors, could make a big difference to recoverable reserves, prolonging the production life of producing fields and postponing the peak of conventional oil production: a mere one percentage point increase in the average recovery factor at existing fields alone would add more than 80 billion barrels (two years of current consumption) or 6% to the world’s proven oil reserves.
The total, long-term potentially recoverable oil resource base, including extraheavy oil, oil sands and oil shales, is estimated at around 6.5 trillion barrels. Adding coal-to-liquids and gas-to-liquids increases this potential to about 9 trillion barrels. Of these totals, nearly 1.1 trillion barrels have already been produced, for the most part at a cost of up to $30 per barrel in year-2008 dollars (excluding taxes and royalties). The cost of exploiting remaining conventional resources typically ranges from $10 to $40 per barrel, while exploitation of oil sands costs between $30 and $80. EOR costs vary between $10 and $80 per barrel and oil shales from $50 to well over $100.
Classifying hydrocarbon resources The amount of fossil-hydrocarbon resources (or hydrocarbons initially in place) on the earth is finite. These resources can be categorised according to the degree of certainty that they exist and by the likelihood that they can be extracted profitably. Several different protocols exist for classifying resources, many of them developed by national organisations.1 This has caused confusion and inconsistency in measuring and comparing resources.
Under the PRMS and UNFC systems, “proved reserves” (or 1P reserves) are hydrocarbons that have been discovered and for which there is a 90% probability that they can be extracted profitably (on the basis of assumptions about cost, geology, technology, marketability and future prices). “Proved and probable reserves” (2P reserves) include additional volumes that are thought to exist in accumulations that have been discovered and that are expected to be commercial, but with only a 50% probability that they can be produced profitably. “Possible reserves” (3P reserves) add to the 2P reserves volumes in discovered fields with a 10% probability of being profitable. Estimates of reserves in each category change over time as the underlying assumptions are modified or new information becomes available. 1. See Etherington and Ritter (2007) for a detailed outline.
198
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
Efforts have been made to achieve a harmonised international approach. The Society of Petroleum Engineers (SPE), the World Petroleum Council (WPC), the American Association of Petroleum Geologists (AAPG) and the Society of Petroleum Evaluation Engineers (SPEE) published jointly in 2007 guidelines on resource definition and classification, called the Petroleum Resources Management System (PRMS). This system is compatible with the 2004 UN Framework Classification for Fossil Energy and Mineral Resources (UNFC), which has been developed by the UN Economic Commission for Europe (UNECE). The PRMS classifies resources and reserves according to the level of certainty about recoverable volumes and the likelihood that they can be exploited profitably (Figure 9.1). The main innovation concerns a shift to a projectbased methodology whereby estimated recoverable quantities are related to actual investment. The classification applies to both conventional and non-conventional resources.
Deterministic and probabilistic methodologies are used to estimate reserves. Reservoir engineers typically apply a number of different formulas involving a range of variables and probabilities to test the sensitivity of a preferred approach or set of assumptions. Increasingly, companies rely on seismic data to map out hydrocarbon reservoirs. But the US Securities and Exchange Commission (SEC), which imposes reserves-reporting standards for companies quoted on US stock exchanges, until now has not allowed companies to book reserves on undrilled acreage on the basis of seismic data alone: it also requires evidence from drilling. The price of oil or gas is a key parameter for assessing the potential profitability of reserves. As prices rise, some reserves that were previously classified as non-commercial may become potentially profitable and be moved into the possible, probable or proven category.
Proved (1P)
Proved + probable (2P)
Proved + probable + possible (3P)
Low estimate
Best estimate
High estimate
Unrecoverable PROSPECTIVE RESOURCES Low estimate
Best estimate
High estimate
Increasing degree of geologic assurance and economic feasibility
Commercial
PRODUCTION
CONTINGENT RESOURCES
Subcommercial
Discovered petroleum initially in place
Hydrocarbon resource classification
Undiscovered petroleum initially in place
Total petroleum initially in place
Figure 9.1
9
Unrecoverable
While progress has been made in establishing a harmonised system of defining and classifying resources, the way those resources are measured in practice still differs widely by country and by jurisdiction. There is no internationally agreed benchmark or legal standard as to how much proof is needed to demonstrate the existence of a discovery, nor about the assumptions to be used to determine whether discovered oil can be produced profitably. This is, in part, because different reporting systems are designed for different purposes. Standards for financial reporting, such as SEC rules, are typically the strictest and result in the lowest estimates. In addition, the extent of the requirements for companies to release information on resources and reserves differs markedly. Auditing reserves and publishing the results are far from universal practice. Many oil companies, including international oil companies, use external auditors and publish the results, but most national oil companies do not. Chapter 9 - Turning oil resources into reserves
199
© OECD/IEA, 2008
Source: SPE/WPC/AAPG (2007).
These inconsistencies have created confusion over just how much oil can reasonably be expected to be produced commercially in the long term. Box 9.1 explains how resources are defined and measured in the Outlook.
Box 9.1
Defining and measuring conventional and non-conventional hydrocarbon reserves in the WEO
A number of bodies, including the IEA, UNECE, SPE, WPC, AAPG and the SEC, are working together to try to harmonise the way in which different types of reserves are measured in practice, with the aim of bringing transparency to resource reporting. In particular, there is a need to ensure that financial and accounting standards for hydrocarbon exploration and production are consistent with business and technical practice. In recognition of this, the SEC has itself proposed new rules on reserves disclosure (Box 9.2), and efforts are being made to ensure that harmonised rules on reporting and communicating reserves estimates are implemented as widely as possible. Progress is hampered by the reluctance to accept new standards on the part of countries and industries that have developed their own reporting systems, and difficulties in adapting national systems to a universal system. Some appreciable progress has been achieved in data transparency on oil production, thanks to the Joint Oil Data Initiative (JODI). A similar approach could be considered to improve the quality of the data on reserves. A successful effort to address both concerns would contribute to better decision making by all interested parties. 200
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
The PRMS classification applies to both conventional and non-conventional resources. Unless otherwise stated, in this analysis, non-conventional oil includes extra-heavy oil, oil sands, oil shales, coal-to-liquids and gas-toliquids. It does not include vegetable oil derived from biomass. But there is also a category of oil which, though considered conventional, can be extracted only using new technologies which are not yet fully developed and/or involves pioneering work in frontier areas, such as ultra-deepwater. Included in this category is oil recovered from as-yet unopened areas of the Arctic and from new deepwater and ultra-deepwater resources, together with additional oil recovered from new enhanced oil recovery projects. There are very large uncertainties about the extent of economically recoverable resources in this category and this “conventional oil produced by unconventional means”, which is discussed below, is often not included in estimates of ultimately recoverable resources. For example, it is not included in the latest figures from the US Geological Survey. We make no systematic estimates for each of the various components of this category, but Figure 9.10 in the last section of this chapter gives a broad illustration of their potential size. Additional oil resulting from reserves growth is customarily included in conventional oil, except where the reserves growth derives from the new EOR projects mentioned above.
Box 9.2
Proposed new SEC reserves-reporting rules
In July 2008, the US Securities and Exchange Commission announced proposed revised oil and gas company reporting requirements aimed at providing investors with a more accurate and useful picture of the oil and gas reserves that a company holds. Reflecting the significant changes that have taken place in the oil and gas industry since the adoption of the original reporting requirements more than 25 years ago, the SEC proposal permits companies to take into account improved technologies and alternative extraction methods, enabling them to provide investors with additional information about their reserves. Specific changes to the rules include the following: Allowing for the use of new technologies to determine which reserves may be classified as proved, if those technologies have been demonstrated empirically to lead to reliable conclusions about reserves volumes. Permitting companies to disclose their probable and possible reserves to investors. Current rules limit disclosure to proved reserves only. Allowing previously excluded resources, such as mineable oil sands, to be classified as oil and gas reserves. Currently these resources are considered to be mining reserves.
9
Requiring companies to report the independence and qualifications of the assessor or auditor, based on the criteria of the Society of Petroleum Engineers. Where companies rely on a third party to prepare reserves estimates or conduct a reserves audit, requiring these reports to be filed. Requiring companies to classify and report oil and gas reserves using an average price based upon the prior 12-month period, rather than yearend prices, to maximise the comparability of reserve estimates among companies by eliminating the distortion of the estimates that can arise through the use of a single pricing date. Source: SEC (www.sec.gov/rules/proposed/2008/33-8935.pdf).
Latest estimates of conventional oil reserves and resources Various organisations compile and report data on oil and gas reserves, using national and company sources. The most widely quoted primary sources of global proven oil reserves (1P) data are the Oil and Gas Journal (O&GJ) and the World Oil Journal. OPEC compiles data for its member countries, and publishes this together with data for other countries drawn from BP. IHS compiles data for 2P reserves only. Other organisations, including BP, publish their own global estimates based mainly on data from primary sources. Chapter 9 - Turning oil resources into reserves
201
© OECD/IEA, 2008
Proven reserves
Current estimates of remaining economically recoverable reserves of oil worldwide do not vary greatly, despite differences in the way they are reported. O&GJ puts global proven oil reserves (including natural gas liquids, NGLs) at the end of 2007 at 1 322 billion barrels, while World Oil, at the other end of the spectrum, reports 1 120 billion barrels (Figure 9.2). The IHS estimate is 1 241 billion barrels, implying a lower figure for proven, 1P reserves. The O&GJ estimate includes 173 billion barrels of Canadian oil sands. The totals reported by BP,2 IHS and World Oil exclude Canadian oil sands; World Oil also excludes NGLs. Figure 9.2
Estimated remaining world oil reserves, end-2007
World Oil OPEC WEC BP IHS O&GJ 1 000
1 050
1 100
1 150
1 200
1 250
1 300
1 350 Billion barrels
Note: All estimates are for 1P reserves, except the IHS estimate which is for 2P reserves. Oil includes natural gas liquids in all cases except World Oil. BP, O&GJ and OPEC include oil sands in Canada (BP excludes reserves that are not under active development). The WEC and World Oil estimates are for end-2005. Sources: BP (2008); O&GJ (2008); OPEC (2008); WEC (2007); World Oil (2006); IHS database.
Globally, remaining proven reserves have increased modestly but steadily since 1990, despite continuing consumption growth. According to O&GJ, reserves increased by 14.2 billion barrels, or 1.2%, in 2007, mainly thanks to upgrades in Venezuela, Brazil, Saudi Arabia, Kuwait, Iran and Angola. OPEC countries’ reserves, at 930 billion barrels based on official data, represent 77% of the world total (using the O&GJ estimate). The global reserves-to-production ratio (R/P), based on current levels of production, is in the range of 40 to 45 years, depending on the source and whether non-conventional resources are included. This level has changed little in the last few years. 2. In its last two Statistical Reviews, BP has included separate estimates of Canadian oil sands. Those reserves are estimated at 152.2 billion barrels at end-2007 (BP, 2008).
202
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
According to BP, which compiles published official figures, proven reserves worldwide have almost doubled since 1980 (Figure 9.3). Most of the changes result from increases in official figures from OPEC countries, mainly in the Middle East, as a result of large upward revisions in 1986-1987. They were driven by negotiations at that time over production quotas and have little to do with the discovery of new reserves or physical appraisal work on discovered fields. The official reserves of OPEC countries have hardly changed since then, despite ongoing production.
Billion barrels
Figure 9.3
Proven remaining oil reserves by region, 1980-2007 (end-year)
1 400
Asia/Pacific
1 200
North America
1 000
Latin America
800 Africa 600 Europe and Eurasia 400 Middle East
200 0
1980
1990
2000
2007
Source: BP (2008).
A growing share of the additions to reserves has been coming from revisions to estimates of the reserves in fields already in production or undergoing appraisal (reserves growth) rather than from new discoveries (see the next section for a discussion of the role of reserves growth). The volume of oil in new discoveries has fallen well below the volume of oil produced in recent years (Figure 9.4). Discoveries dropped from an average of 56 billion barrels per year in the 1960s to 13 billion barrels per year in the 1990s. The number of discoveries peaked in the 1980s, but it fell sharply in the 1990s. The fall in both the number of discoveries and the volume of oil discovered has been most acute in the Middle East and the former Soviet Union. By contrast, discoveries rose in Africa, Latin America and Asia. The drop in discoveries worldwide was largely the result of a fall in exploration activity in the regions with the largest reserves (where access by international companies is most difficult) and a fall in the average size of the fields that have been found. Oil discoveries* and production, 1960-2006
60
1 600 1 400
50
1 200 40
1 000
30
800
Billion barrels
Billion barrels per year
Figure 9.4
9
Discoveries Production Cumulative discoveries (right axis) Cumulative production (right axis)
600
20
400 10
0 1960-1969
1970-1979
1980-1989
1990-1999
2000-2006
* Additions to proven reserves from new fields. Sources: IHS and IEA databases.
Chapter 9 - Turning oil resources into reserves
203
© OECD/IEA, 2008
200
0
These factors more than offset the effect of an improved exploration drilling success rate, which has increased by a factor of two in the last 50 years, from one out of six wildcat wells to one out of three today (Harper, 2005). The downward trend in the volume of oil found has reversed slightly in the current decade, thanks to increased exploration activity (with higher oil prices), with an average 16.4 billion barrels per year discovered since 2000. The excess of production over cumulative discoveries over the same period is 400 billion barrels.
Reserves estimates give an indication of how much oil could be developed and produced in the near to medium term. The total volume of oil which might ultimately be produced commercially is known as ultimately recoverable resources. This category includes oil initial proven and probable reserves from discovered fields (including oil already produced), both in production and yet to be developed, reserves growth (see below) and hydrocarbons that are yet to be found. From a larger resource base, more oil is likely to progress into the proven category, deferring the peak in global production and permitting more oil to be produced. The US Geological Survey, a federal government agency, is the world’s leading source of estimates of recoverable hydrocarbon resources. It carried out a major assessment of the world’s conventional oil and gas resources in 2000, based on raw data for 1995 (USGS, 2000). Since that time, it has reassessed some basins and assessed some others for the first time. Our current estimates of initial and remaining ultimately recoverable conventional resources are shown in Table 9.1. These take account of the USGS updates, new remaining reserve estimates from IHS and cumulative oil production to the end of 2007, but exclude the “conventional oil produced by unconventional means” discussed later in Box 9.3. On this basis, using the mean figures of the USGS, the world’s current ultimate recoverable conventional oil and NGL resources are estimated to amount to just under 3.6 trillion barrels. Of these, cumulative production to date is 1.1 trillion barrels. Remaining resources amount to over 2.4 trillion barrels. Reserves growth accounts for around 400 billion barrels, with the largest contribution to reserves growth (in proportion to the reserves) over the last decade from the Middle East, Latin America and Africa. Remaining recoverable resources are largest in the Middle East, Russia and the Caspian region. Despite the thoroughness of the assessments, there are considerable uncertainties surrounding the estimates of ultimate recoverable resources — a point highlighted by the USGS itself. In addition to the mean value mentioned above, the USGS provides alternative numbers based on different probabilities, ranging from 2 400 billion barrels with a 90% probability of at least that amount being available to 4 495 billion barrels with a 5% probability (not adjusted for reserves growth since the 2000 assessment was released). The USGS recently reviewed trends in discoveries and reserves growth in the period 1996-2003, comparing them with their original estimates of the long-term potential (Klett et al., 2007). It found that, based on IHS data, 28% of the estimated additions (mean value) to 2P oil reserves from reserves growth (to 2025) had already been achieved — almost exactly in line with the pace of growth predicted in the 2000 assessment (Figure 9.5). The rate of discovery lagged this figure — 11 % of estimated undiscovered oil volumes had been discovered by 2003 — but the ultimate timescales 204
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
Ultimately recoverable resources
are not directly comparable. Overall, 71% of the total additions to reserves came from reserves growth and only 29% from discoveries. In all areas except for sub-Saharan Africa and Asia-Pacific, reserves growth outpaced discoveries. For natural gas, a higher proportion of estimated potential from reserves growth had actually been attained (51%) but a smaller share of the potential from discoveries (less than 10%). The share of reserves growth in the total reserve additions between 1996 and 2003 was 79%. Table 9.1
OECD North America Europe Pacific Non-OECD E. Europe/Eurasia Asia Middle East Africa Latin America World
World ultimately recoverable conventional oil and NGL resources, end-2007 (mean estimates, billion barrels) Initial reserves
Cumulative production
Reserves growth
[1]
[2]
[3]
458 368 77 13 1 911 355 134 986 206 229 2 369
363 300 56 7 765 171 79 312 102 100 1 128
27 22 3 2 375 67 20 204 40 44 402
Undiscovered Ultimately Remaining Remaining resources recoverable reserves recoverable resources resources [4] [5 = 1+3+4] [6 = 1-2] [7 = 5-2] 185 95 80 10 620 140 30 257 85 108 805
670 485 160 25 2 907 562 184 1 447 331 381 3 577
95 68 20 6 1 147 184 55 674 104 129 1 241
307 185 103 18 2 142 391 105 1 135 229 281 2 449
9
Sources: USGS (2000); Klett et al. (2007); USGS data provided to IEA; IHS databases; IEA databases and analysis.
Billion barrels of oil equivalent
Figure 9.5
New field discoveries and reserves growth in 1996-2003 compared with USGS 2000 assessments
800
2000 assessment Actual 1996-2003
600
400
200
0 Discoveries Oil
Reserves growth
Discoveries Gas
Note: Reserves growth potential in the 2000 USGS assessment is to 2025, while discoveries relate to ultimate resources. Source: Klett et al. (2007).
Chapter 9 - Turning oil resources into reserves
205
© OECD/IEA, 2008
Reserves growth
Arctic resources The USGS continues to assess global hydrocarbon resources, focusing on particular regions that it judges to be worthy of more in-depth study. In 2005 it launched a major reassessment of the Arctic region (Circum-Arctic Resource Appraisal, CARA), and released the final results in July 2008. Earlier USGS estimates of undiscovered oil and gas in the Arctic, amounting to 25% of total undiscovered resources worldwide, actually included continental areas in Western Siberia below the Arctic Circle. Box 9.3
Potential for exploiting Arctic resources
Oil and gas fields in the arctic frontier area, north of the Arctic Circle, currently account for a small share of the world’s hydrocarbon production. But that share could rise significantly in the coming decades as technical challenges are addressed and costs lowered. The Arctic presents many of the high-profile challenges associated with deepwater operating conditions: remoteness, personnel safety, environmental footprints and high costs. An extreme climate and hazards from ice and icebergs add to these difficulties. Significant knowledge of operations in the Arctic was gained in the 1960s and 1970s in Russia, Canada and Alaska. More recently, developments in the Yamal Peninsula, the Snohvit LNG project in the Barents Sea and the offshore Sakhalin projects in Russia, in Arctic-like conditions, have added to that experience. Some prospects are near existing infrastructures such as the North Slope of Alaska and the Pechora Sea in Russia. Other basins are far from areas that have already been explored or where resources have already been exploited.
Estimates of exploration and production costs in the Arctic vary widely, mainly as a function of the distance from existing fields and the nature of the operating environment. Indicative costs for an offshore arctic production field, including drilling, production facilities, and operating and decommissioning components, are currently about $35 to $40 per barrel for the easiest resources to exploit.
206
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
The ability to drill wells in ice-covered seas will be critical to the efficient development of Arctic resources. There is generally a limited operational window for existing drilling units in ice-covered waters, though the receding ice-cap will increase the size of this window. Fit-for-purpose arctic drillships are being developed in Norway for water depths of more than 80 metres. New technologies such as subsea and sub-ice drilling are being considered. A majority of offshore fields will most likely be developed using subsea technology and well stream transport to land, as was the case with the Snohvit field in Norway (Johnsen, 2008). For easing oil flow at the surface, effective solutions are still to be developed, including fluid conditioning, pressure-boosting and wellintervention technologies. Oil transport by ships is typically done by tankers with a capacity of 100 000 barrels, towed by icebreakers. A tanker being developed by StatoilHydro will be able to carry more than 500 000 barrels of oil (or half as much as a standard very large crude carrier) without the need for icebreakers, except where there are large ice ridges.
The CARA (USGS, 2008) study covers 31 geological provinces, of which 25 are considered to hold hydrocarbon potential, and 69 assessment units. The new study shows increased potential gas resources — 1 669 trillion cubic feet of gas (30% of the world’s undiscovered conventional gas) — but lower quantities of oil — 90 billion barrels (mean value) of undiscovered conventional oil and 44 billion barrels of NGLs (13% of the world’s total undiscovered conventional liquids). Deepwater and ultra-deepwater resources Estimates of the recoverable conventional oil located in deepwater (water depths between 400 meters and 1 500 metres) and ultra-deepwater (water depths greater than 1 500 meters) lie between 160 and 300 billion barrels (IEA, 2005). Over 70% of these resources are in Brazil, Angola, Nigeria and the United States. Most production to date has occurred in the US Gulf of Mexico, but production is rising in Angola and Brazil (where several large discoveries have been made in the stratum below the salt deposits, the presalt layer, Box 9.4). New developments are underway in Mexico, Nigeria, Southeast Asia and the Mediterranean Sea. Other areas of deepwater exploration interest include the UK Atlantic margin, the Black Sea, India and the Arctic. Cumulative global oil discoveries in deepwater offshore areas, plotted against the number of exploration wells, remain on
9 Box 9.4
Pre-salt deepwater potential in Brazil
A host of technical difficulties need to be overcome for these pre-salt finds to be developed: the complexity of drilling deviated and horizontal wells through the salt zones, slow drilling rates and the need, in some cases, to hydraulically fracture wells. Secondary recovery, using water or alternated water-gas injection, creates additional challenges. Finally, once the wells start producing, flow assurance requires a careful strategy for preventing the deposition of paraffins and other substances in production lines. Adding to the complexities are the handling of associated natural gas and CO2, and high pressures.
Chapter 9 - Turning oil resources into reserves
207
© OECD/IEA, 2008
The Santos Basin offshore Brazil is one of the most promising regions for deepwater production. Several discoveries have been made in the pre-salt (or sub-salt) section of the Basin. These include the Tupi and Jupiter condensate/ gas fields in 2007, each with estimated recoverable reserves of up to 8 billion barrels of oil equivalent, and, in 2008, the Carioca field. Brazil’s proven oil and gas reserves are expected to be revised sharply upwards once these and other new finds have been appraised. Development plans have already been submitted for some of them. Production testing on Tupi, located in more than 2 000 metres of water and 4 000 metres below the seabed, is due to begin in 2009. Petrobras, which holds a 65% stake in the field, plans pilot production of around 100 thousand barrels per day (kb/d) in 2011-2012 and at least 200 kb/d within 10 to 15 years. Peak output of as much as 1 mb/d is possible. Depending on other deepwater finds, the Santos Basin could contribute the bulk of the increase in Brazil’s total oil production over the projection period.
a linear trend (Figure 9.6), suggesting that more large finds, such as the recent Brazilian Tupi field, are likely to be made in the future. Future prospects are classified here as conventional oil produced by unconventional means.
Cumulative discoveries (billion barrels)
Figure 9.6
Global offshore oil discoveries to end-2006 versus exploration wells
500 450 400 350 300 250 200 150 100 50 0 0
5 000
10 000
15 000 20 000 Cumulative number of exploration wells
Source: Sandrea and Sandrea (2007a).
Nonetheless, wells drilled in ultra-deepwater locations face major technological challenges, including high underground temperatures and pressures (which require special metals), the behaviour of salt canopies (which can be more than 3 000 metres thick) and near-surface gas zones. In addition, complex hydraulic fracturing may be needed to maintain the structure of the reservoir during production and to avoid the in-flow of sand (Couvillion, 2008). The design of sub-sea production systems needs to evolve to provide assurance of the unimpeded flow of hydrocarbons and to extend the production life of fields. 208
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
The oil and gas industry has made great strides in developing technologies to explore for and produce oil and gas in ever deeper water locations. Total world production from deepwater fields rose from less than 200 kb/d in 1995 to over 5 million barrels per day (mb/d) in 2007. Production from ultra-deepwater fields began only in 2004 and is expected to exceed 200 kb/d by 2010. The average water depth of production wells has more than tripled since the mid-1990s, technical advances having enormously increased the maximum water depth in which wells can be drilled. The 3 000 metres water depth mark for wells drilled was exceeded by the recent Chevron Toledo project in the Gulf of Mexico. A record for the combined depth of water and the well itself was set by another Chevron project, called Knotty Head, also in the Gulf of Mexico. In addition, the time needed to bring deepwater fields into commercial production is shortening (Figure 9.7). Improvements in seismic data gathering and processing technologies that enable images to be prepared of areas located beneath thick layers of salt zones are raising the discovery rate.
Metres
Figure 9.7
Maximum water depth of offshore exploration and production wells worldwide
3 500
Exploration
3 000
Production
2 500 2 000 1 500 1 000 500 0 1965-1969 1970-1974 1975-1979 1980-1984 1985-1989 1990-1994 1995-1999 2000-2004 2005-2009 Source: Karlsen (2008).
Although there is thought to be a considerable amount of oil in place in deepwater locations, how much is recoverable will hinge on the balance between development costs and oil prices (as of course, is true of all oil reserves, onshore and offshore). Strong demand for deepwater drilling rigs (semi-submersible platform and drillships, collectively referred to as floaters) has driven up rig rates in recent years (see Chapter 13). At mid-2008, there was a near-record tally of floaters under construction, comprising 44 semi-submersibles and 30 drillships. Most of the new floaters will be capable of drilling in water depths greater than 7 000 feet (2 300 metres), and half of them could drill in more than 10 000 ft (3 300 metres) of water. The majority of the more than 100 offshore rigs built in the 1970s, which are expected to be retired in the next five years, are capable of drilling only in shallow water (less than 400 metres). Investment in deepwater drilling is expected to continue to grow: from $58 billion in 2001-2005 to $108 billion in 2008-2012, with 85% going to the “golden triangle” of North America, Brazil and West Africa, according to recent industry forecasts (DouglasWestwood, 2007). Asia accounts for a further 10%, with activity in Indonesia, India and Malaysia. Sub-sea wells represent the largest components of capital costs (41%), followed by platforms (32%), pipelines and control lines (16%) and sub-sea hardware (9%). At present, a minimum oil price of $45 in the Gulf of Mexico and $60 in Angola is needed to yield a 12% return on investment.
9
Reserves growth, which refers to the increase in recoverable reserves from a discovered oil or gas field that occurs over the life of the field as it is appraised, developed and produced, is expected to be a major contributor to additions to reserves in the coming decades. It derives from three factors (Klett, 2004): Chapter 9 - Turning oil resources into reserves
209
© OECD/IEA, 2008
Reserves growth and enhanced oil recovery
Geological factors: The delineation of additional reserves through new seismic acquisition, appraisal drilling or the identification (using well-bore measurements), of reservoirs that had been previously bypassed. Technological factors: An increase in the share of oil in place that can be recovered through the application of new technologies, such as increased reservoir contact, improved secondary recovery and enhanced oil recovery. Definitional factors: Economic, logistical and political/regulatory/fiscal changes in the operating environment. The Weyburn field in Canada provides a good illustration of how reserves can grow over time (Figure 9.8). Production started in the mid-1950s with conventional vertical wells. Initial proven reserves at that time were estimated at 260 million barrels. Infill drilling of wells in the mid-1980s added a producible wedge of 20 million barrels. Horizontal wells drilled in the 1990s added a second, additional wedge of reserves of around 60 million barrels. At the end of the 1990s, an enhanced oil recovery scheme — using anthropogenic CO2 piped in from a coal-gasification plant in North Dakota in the United States — added 120 million barrels more, or one-third, taking initial reserves to 460 million barrels.
Thousand barrels per day
Figure 9.8
A case study of oil reserves growth: the impact of technology on oil production from the Weyburn field in Canada
50
CO2 injection Infill - horizontal wells
40
Infill - vertical wells Original vertical wells
30 20 10 0 1950
1960
1970
1980
1990
2000
2010
2020
2030
Source: PTRC Weyburn-Midale website (www.ptrc.ca).
Reserves growth in any given field automatically raises the recovery factor, which is defined as total recoverable reserves (including the oil or gas already produced) expressed as a percentage share of the original hydrocarbons in place. As estimates of both the volume of hydrocarbons in place and how much is recoverable vary as a field is developed and produced, the estimated recovery factor inevitably fluctuates. 210
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
The USGS has undertaken a comprehensive analysis of the main factors in reserves growth, based on production series for 186 giant fields outside the United States (133 in OPEC countries) with more than 0.5 billion barrels of reserves for the period 1981 to 2003 (Klett, 2004). Reserves in the OPEC fields increased by 80% in aggregate and in non-OPEC fields by around 30% over this period.
Consequently, estimating trends in recovery factors is extremely difficult. The global average recovery factor was recently estimated at 34.5%, based on IHS data (Schulte, 2005). Fields containing more than two-thirds of conventional oil reserves have a recovery factor of less than 40%. Lifting recovery factors could make a big difference to recoverable reserves, prolonging the production life of producing fields and postponing the peak of conventional oil production: a mere one percentage point increase in the average recovery factor would add more than 80 billion barrels (or 5% to 6%) to the world’s proven oil reserves (see Spotlight below). Enhanced (or tertiary) oil recovery (EOR) technologies are applied to increase the amount of oil recovered from a reservoir beyond the range that primary and secondary recovery can generally achieve (generally between 20% and 50%). EOR is expected to contribute significantly to future reserves growth. The technologies utilised in EOR aim at altering the properties of the oil and improving its movement from the reservoir to the wellhead. EOR currently accounts for only about 3% of world oil production. There are three main types of EOR techniques in use today: thermal recovery, miscible displacement (using nitrogen, CO2 or other hydrocarbons — natural gas in the case of EOR), and chemical flooding (with micellar-polymers or alkalines). Other technologies include microbial EOR and foam flooding.
9
Research and development on EOR technologies, especially those involving miscible displacement and CO2 injection, which started before 1950, intensified after the first oil shock in the early 1970s. This effort, along with the development of expertise in candidate reservoir screening, EOR project implementation and field management, resulted in the deployment of many of the projects that are currently in operation today, most of which are in the United States. But low oil prices in the 1990s led to a significant loss of expertise amongst oil companies and technology providers. As a result, a renewed effort is needed to improve the prospects for expanding the deployment of EOR in the longer term, particularly in the following areas (NPC, 2007): Monitoring and controlling the injection of CO2. Complex multi-lateral wells with control of individual well-branch oil inflow. Optimisation of steam-assisted gravity drainage (SAGD). Reservoir-scale measurements, including 4-D seismic.
Additional production resulting from the use of EOR techniques currently amounts to about 2.5 mb/d, with reliance mainly on thermal processes (in the United States, Canada, Venezuela, Indonesia and China), CO2 injection (United States), natural gas injection (Venezuela), nitrogen (in the Cantarell field in Mexico) and polymer/ surfactant flooding (China) (Table 9.2). In the Reference Scenario, additional production resulting from EOR techniques (over and above current EOR output) is projected to rise to over 6 mb/d by 2030, with EOR using CO2 (CO2-EOR) accounting for most of the increase (see Chapter 11). Worldwide, the prospects for CO2-EOR, which can be costly, have been boosted by higher oil prices (Box 9.5). Chapter 9 - Turning oil resources into reserves
211
© OECD/IEA, 2008
Smart oil lifting, with re-injection of water in the same well.
S P O T L I G H T
A 2007 study by the US National Petroleum Council lists advances in technology that offer the potential to improve oil recovery factors. It estimates that the biggest impact would come from technologies that maximise the effective reservoir contact between the well bore and the formation through various techniques, such as controlled multi-lateral drain holes, improvements in the efficiency of secondary and tertiary recovery, better imaging of reservoirs and production management in real-time, and the increased use of 4-D seismic to monitor oil flows, and “smart” injector and producer wells. The commercialisation of some of those technologies may not happen until the end of next decade, but operators are already conducting full-field experiments that could advance the time scale. In the North Sea’s Norne field, the operator StatoilHydro has used new seismic acquisition and processing technologies, and combined them with geological and reservoir models and well-bore measurements to monitor fluid movements across the field (Boutte, 2007). These techniques helped to improve the planning of in-field drilling, to increase the field recovery factor from 40% to 52%, and to extend the life of the field to beyond 2015. Technologies that improve contact between the well bore and the reservoir show great promise. In Saudi Arabia, the maximum reservoir contact (MRC) approach, used in the development of the Shaybah field, demonstrated that well productivity could be increased significantly by drilling long horizontal and multilateral wells. By increasing the length of well in contact with productive areas from 1 km to 12 km, through use of multi-lateral wells, well productivity was increased by a factor of six (Saleri et al., 2003). However, only a limited number of lateral branches can be extended from each mother bore and it is impossible to control the flow from individual sections. This can cause a problem as water production rises. Extreme reservoir contact (ERC), a concept being developed by Saudi Aramco, aims to get around this problem. ERC wells improve on MRC through the development and deployment of new intelligent well completions that improve recovery by means of downhole monitoring and electrically activated downhole control valves, without the use of complex control lines that run from the wellhead. The ERC concept could allow an unlimited number of lateral branches and boost the recovery from heterogeneous carbonate reservoirs. However, the widespread introduction of technologies like these that increase reservoir contact would require a significant increase in footage drilled, which would call for more drilling rigs and personnel — both of which are in short supply at present (see Chapter 13). It will probably take much more than two decades for the average recovery factor worldwide to be raised from about 35% today to 50%. Achieving this would boost world reserves by about 1.2 trillion barrels — equal to the whole of today’s proven reserves (Figure 9.3).
212
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
How soon – if ever – could oil recovery factors be raised to 50%?
Table 9.2
Estimated 2007 incremental production from EOR projects by technology (thousand barrels per day) Thermal
Nitrogen
CO2/
Chemical
Total
hydrocarbon United States
300
10
340
–
650
Canada
330
–
50
–
380
Mexico
–
500
–
–
500
Venezuela
200
–
170
–
370
China
160
–
5
200
365
Indonesia
190
–
–
–
190
Others
20
–
35
–
55
World
1 200
510
600
200
2 510
Note: Incremental production rate is calculated by deducting the projected rate without EOR from the enhanced production rate with EOR. Current EOR production is included in total crude oil production from existing fields in the projections in Chapter 11. Sources: Thomas (2008); O&GJ (2008).
Carbon-dioxide enhanced oil recovery (CO2-EOR)
9
CO2-EOR was initially deployed in the 1970s and is still used only onshore, with a small inter-well distance. In addition to expanding the use of the technology to new onshore areas, there is considerable potential for using it offshore and in thicker reservoirs. An assessment of the potential in the United States shows that improved CO2-EOR, in combination with 4-D seismic, advanced well monitoring and control, and reservoir simulation, could add between 20 and 40 billion barrels of economically recoverable oil (ARI, 2006). The worldwide potential for CO2-EOR is of the order of 160 billion to 300 billion barrels, equal to 7% to 14% of current remaining conventional recoverable oil resources. The Middle East and North America are thought to hold the greatest potential (Figure 9.9). The potential is greatest in fields that have relatively low depletion levels, as the costs of installing the equipment and facilities required are onerous, particularly for offshore fields (Tzimas et al., 2005). Incremental recovery factors with EOR from offshore fields are expected to be lower than onshore fields because of different well configurations and reservoir contacts. CO2-EOR is now increasingly Chapter 9 - Turning oil resources into reserves
213
© OECD/IEA, 2008
CO2-EOR, which involves the injection of CO2 into oil reservoirs, is receiving increased attention because of the potential availability of CO2 from anthropogenic sources and the environmental benefits of carbon sequestration. There are currently more than 100 CO2-EOR projects in the world today, most of them in North America, producing 300 kb/d of additional oil. In most cases, the CO2 is extracted from produced natural gas. In the United States alone, the contribution from CO2-EOR has been projected to increase from the current level of 250 kb/d to 1.3 mb/d by 2030 (Caruso, 2007). The economics of CO2-EOR are attractive, as the injection of one tonne of CO2 into a suitable reservoir leads to an incremental recovery of between two and three barrels of oil. The value of this oil could be set against the cost of capturing the CO2 from power plants and transporting it to the injection fields, which is estimated to range from $50 to $100 per tonne (IEA, 2008).
Box 9.5
Deploying EOR: when, how and at what cost?
The applicability of EOR depends on the characteristics of the oilfield and its state of depletion. For a field that is conducive to EOR, there is generally a window of opportunity during which such technology is best applied. Table 9.3 shows the key criteria used for assessing the applicability of EOR technologies. The gravity of the crude oil is generally the first parameter considered: the majority of the miscible flooding techniques work only for light crudes. The second parameter is the amount of oil left in the field. CO2-EOR can be used when more than 20% of the oil is still recoverable. The oilfield temperature is the next variable: polymer flooding (alkaline and micellar) does not work at temperatures over 95°C. Hydrocarbon, polymer and surfactant flooding have the greatest potential for increasing recovery factors. Alberta’s hydrocarbon miscible flooding projects have achieved recovery factors of more than 70%. A massive polymer flooding project combined with surfactant addition in China’s Daqing complex has boosted production by 240 kb/d and the recovery factor by more than 12%. EOR costs have risen sharply in the last few years, in line with upstream costs generally. Energy feedstock prices for thermal processes and hydrocarbon displacement, increased demand for CO2, and the impact of higher oil prices on surfactants and polymers have contributed to these higher costs. On the other hand, EOR costs are generally much lower than the cost of other means of reserves replacement, especially in North America. They range from less than $20 to $70 per incremental barrel of oil, excluding the cost offset that might be available through any CO2 emission credit that might exist.
Pre-screening criteria for EOR
Technology
API
Remaining Formation recoverable type resources (carbonate/ (% of intitial sandstone) reserves)
Nitrogen
>35
>40
Hydrocarbon
>25
>30
Depth (metres)
Permeability (md)
Temperature (°C)
Required additional recovery factor (%)
Carbonate
>2 000
190
–
–
Carbonate
>1 350
–
–
20-40
CO2
>25
>20
Carbonate
>700
–
–
5-25
Polymer
>15
>70
Sandstone
10
18
>35
Sandstone
10
10
>50
Sandstone
>50
>50
>40
–
Thermal/ Steam
>8
>40
Sandstone
200
–
10-60
Sources: Taber et al. (1997); Green (1998); O&GJ (2008); IEA analysis.
214
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
Table 9.3
attractive in the North Sea (OGUK, 2008). While an estimated 15 to 25 years of production life could be added to a number of North Sea fields, some are likely to be dropped from the list of potential candidates by 2010, due to less favourable miscible displacement conditions.
Billion barrels
Figure 9.9
Potential additional recoverable oil resources using CO2-EOR by region
140
High of range
120
Low of range
100 80 60 40 20 0 Middle East
North America
Former Soviet Union
Latin America
Africa
Asia/Pacific
Europe
9
Sources: Khatib (2006); Kuuskraa (2006); IEA estimates based on industry sources.
Non-conventional oil resources Extra-heavy oil and oil sands A large share of the world’s remaining oil resources is classified as non-conventional.3 These resources — oil sands, extra-heavy oil and oil shales — are generally more costly to produce, though considerable progress has been made in addressing technical challenges and lowering costs. Oil sands and extra-heavy oil resources in place worldwide amount to around 6 trillion barrels, of which between 1 and 2 trillion barrels may be ultimately recoverable economically. Technically recoverable reserves, defined by the World Energy Council as resources which can be exploited profitably with today’s technology, amount to some 1.1 trillion barrels using an average conservative recovery factor of 18% (Table 9.4).
3. The IEA defines non-conventional oil as oil sands, extra-heavy oil, oil shales, coal-to-liquids (CTL) and gas-to-liquids (GTL). Only the first four are considered here. The prospects for CTL and GTL are discussed in Chapter 11.
Chapter 9 - Turning oil resources into reserves
215
© OECD/IEA, 2008
The world’s extra-heavy oil and oil sands resources are largely concentrated in Canada (mainly in the province of Alberta) and Venezuela (in the Orinoco Belt). Assuming a potential 20% recovery factor, these two countries would hold more recoverable resources than all the conventional reserves in the Middle East. Alberta alone has proven reserves today of 174 billion barrels (crude bitumen) and an estimated 315 billion barrels of ultimately economically recoverable resources. Nearly 80% of the
proven reserves in Alberta are in the Athabasca Wabiskaw McMurray deposit, while the remaining 20% are split between the Cold Lake Clearwater and the Peace River deposits. A one percentage-point increase in the recovery factor would boost recoverable resources by 60 billion barrels. Table 9.4
Extra-heavy oil and oil sands resources (billion barrels) Extra-heavy oil Oil in place
Recovery factor
Oil sands Technically recoverable
Oil in place
Recovery factor
Technically recoverable
North America
184
0.19
35
1 659
0.32
531.0
South America
2 046
0.13
266
1
0.01
0.1
650
0.12
78
0
0.00
0.0
Middle East Other regions World
423
0.13
55
1 135
0.13
120.0
3 303
0.13
434
2 795
0.23
651.0
Source: WEC (2007).
Several technologies exist to extract bitumen from oil sands. If the deposit is near the surface, oil sands are mined, using enormous power shovels and dump trucks. The bitumen is then extracted using hot water and caustic soda. The extracted bitumen needs to be upgraded (or diluted) with lighter hydrocarbons before it can be transported to a refinery. Upgrading consists of increasing the ratio of hydrogen to carbon, either by carbon removal (coking) or by adding hydrogen (hydrocracking), resulting in a synthetic crude oil that is shipped to a refinery.
Much higher recovery factors (up to twice those of conventional production technologies) can be obtained using in situ viscosity-reduction techniques. In situ technologies are currently used for highly viscous oils: they include cyclic steam stimulation injection (CSS) and steam-assisted gravity drainage (SAGD). Technologies being developed include a vapour extraction process, which uses hydrocarbon solvents instead of steam to increase oil mobility, the use of downhole heaters and hybrid methods. Theoretical recovery factors of 50% to 70% are predicted for SAGD and new in situ processes, significantly higher than with CSS (20% to 35%). The steam-oil ratio is also lower with SAGD (2 to 3) than with CSS (3 to 5), so less water would be required. About 20% of Alberta’s bitumen can be mined, while 80% requires in situ recovery methods (ACR, 2004). 216
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
When the heavy hydrocarbon deposit is deeper, drilling is required. When its viscosity is low enough or can be reduced enough for the oil to flow to the surface, long horizontal or multi-lateral wells are used to maximise well bore contact with the reservoir and reduce the drop in pressure in the well bore. This is the technique used for several of the heavy oil deposits in Venezuela’s Orinoco Belt. The main drawback of such conventional production techniques is the low recovery factors, typically less than 15%. As a result, the expected amount of extra-heavy oil that could be recovered from the estimated 1 700 billion barrels in place in the Orinoco Belt is less than 250 billion barrels, using current technology.
Oil shales: a technological milestone in sight? Oil shales are rocks that contain a large proportion of solid organic compounds (kerogen) and are found at shallow depths, from surface outcrops to 1 000 metres below ground. Production of shale oil dates back to the 1830s and peaked worldwide in 1980 at around 45 000 tonnes. It fell to only 16 000 tonnes in 2004, with more than 70% coming from Estonia. While resources in place are estimated at 2.5 to 3 trillion barrels, estimates of recoverable reserves vary markedly (see, for example, NPC, 2007 and WEC, 2007). In calculating the long-term, oil-supply cost curve (see below) we have taken a figure of about 1 trillion barrels. The United States has the largest resources (more than 60% of the world total), followed by Brazil, Jordan, Morocco and Russia. Oil shales are not projected to make a significant contribution to world oil supply before 2030, but technological breakthroughs could change this outlook. Oil shales near the surface can be mined, with the mined rock heated in a process called retorting, which pyrolyses the kerogen into oil. Retort methods are unlikely to provide an environmentally sustainable production process. Deeper deposits require the use of techniques to enhance the productivity of the formation (such as hydraulic fracturing) and an in situ mobility-enhancement technology. The processes are very energy intensive. Retorting alone requires nearly 30% of the energy value of the oil produced. In addition, CO2 emissions generated are in the range of 0.18 to 0.25 tonnes per barrel of oil produced. Production costs currently range from $50 to $120 per barrel, disregarding any potential carbon permitting.
9
The main US resource is the Green River Formation, with four basins in western states. Early experiments in the 1980s were halted due to the unfavourable economics and poor operational performance. Shell is assessing the technical and economic viability of an in situ conversion process, using down-hole heaters, at the Mahogany Ridge project (IEA, 2007). The oil produced has an API in the mid 30s, significantly higher than oil produced by surface retort. The process has the advantage of reducing the amount of land used, assuming wells can be drilled as little as 12 metres apart, but it requires the use of both water and of a sufficient amount of energy to heat and cool the relevant areas. A 100-kb/d unit would require 1.2 GW of power. Shell expects gas production from the produced fluid to meet most of the energy needs. Preliminary energy balances indicate, on a life-cycle assessment basis, a ratio of 3-4 to 1 between the energy generated and the energy used. Further technical developments to protect groundwater are required before going to a larger scale demonstration phase. Other companies, including Chevron, EGP and IEL, are testing different in situ processes. Shell is not expected to decide on whether or not to proceed with commercialisation of their technology before 2010. If it gets the green light, first production would occur after 2015.
We have updated the oil-supply cost curve, which we first produced three years ago (IEA, 2005),4 based on the latest estimates of resource potential and the estimated range of 4. The IEA will be updating in 2009 the Resources to Reserves study, first published in 2005 (IEA, 2005), with an extended coverage of resources and technologies.
Chapter 9 - Turning oil resources into reserves
217
© OECD/IEA, 2008
Long-term oil-supply cost curve
current costs of production. The curve plots the potential long-term contributions from conventional resources (including those defined above as conventional oil produced by unconventional means) and non-conventional resources (including liquids supply from coal and gas) against their associated current production costs (Figure 9.10). The resource potential is not much different from that estimated in 2005, though technological developments have boosted the recovery potential for some categories of resources, notably ultra-deepwater, EOR and oil sands. But costs have been revised upwards significantly, reflecting the recent surge in costs (see Chapter 13). Long-term oil-supply cost curve
140 Deepwater and ultra deepwater
120 100
Oil shales
60 40 20
Produced
MENA
EOR Arctic
80
CO2 – EOR
Production cost (dollars - 2008)
Figure 9.10
Other convential oil
Gas to liquids
Coal to liquids
Heavy oil and bitumen
0 0
1 000
2 000
3 000
4 000
5 000
6 000
7 000
8 000 9 000 10 000 Resources (billion barrels)
Note: The curve shows the availability of oil resources as a function of the estimated production cost. Cost associated with CO2 emissions is not included. There is also a significant uncertainty on oil shales production cost as the technology is not yet commercial. MENA is the Middle East and North Africa. The shading and overlapping of the gas-to-liquids and coal-to-liquids segments indicates the range of uncertainty surrounding the size of these resources, with 2.4 trillion shown as a best estimate of the likely total potential for the two combined.
Conventional resources, more than half of which are in the Middle East and North Africa, amount to about 2.1 trillion barrels (based on the latest USGS assessments). The cost of exploiting these resources is expected to be much lower on average than the cost of all other sources of supply, with the exception of oil recovered using CO2-EOR. The cost of exploiting conventional resources (excluding taxes and royalties) typically ranges from less than $10 to $40 per barrel, though some fall outside this range. 5. Only a small fraction was produced at a cost of more than $20, mainly in recent years from offshore and North American fields.
218
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
The total long-term potential oil resource base alone is around 6.5 trillion barrels. Adding CTL and GTL increases this potential to about 9 trillion barrels. Of these totals, nearly 1.1 trillion barrels have already been produced at a cost of up to $30 per barrel in year-2008 dollars.5 Remaining potentially economically recoverable resources are broken down as follows:
Additional resources recovered using enhanced oil recovery has been split according to the use CO2-EOR and other EOR technologies. The cost of CO2-EOR ranges from just over $20 per barrel to about $70, compared with about $30 to $80 for other types of EOR. The overall potential from EOR is estimated at between 400 and 500 billion barrels, on the assumption that the introduction of new EOR technologies is accelerated, lowering unit costs. Deepwater and ultra-deepwater resources could deliver over 160 billion barrels at a cost of up to $65 per barrel. Arctic resources could amount to 90 billion barrels at a cost of between $40 and $100 per barrel. Extra-heavy oil and oil sands resources total more than 1 trillion barrels and could be produced at costs ranging from over $40 to about $80 per barrel. Oil shales production costs are estimated to be in the range of $50 to well over $100 per barrel. Because of a lack of major commercial projects, the prospects for improving oil shales production technology are very uncertain. Consequently, we do not expect oil shales to make any significant contribution to world oil supply before 2030. Moreover, a carbon penalty would increase significantly the production cost, as oil shales extraction is extremely energy and carbon intensive. CTL and GTL have a large potential (see also Chapter 11), but they will be in competition for their feedstock with other potential applications (mainly in power generation and final uses). Their costs range from $40 to $120 per barrel at current feedstock prices. Political, environmental, regulatory and fiscal factors will strongly influence the extent to which this potential is exploited and the costs to oil companies of bringing the resources to market. Production of non-conventional resources, in particular oil sands as well as oil shales in the future, leaves a large environmental footprint, including through the greenhouse gases emitted in the production and use. The widespread introduction of CO2 emission-reduction incentives would have a major impact on the cost curve, as non-conventional oil would become relatively more expensive and CO2EOR potentially cheaper. Technological developments will also continue to change the shape and position of the curve.
219
© OECD/IEA, 2008
Chapter 9 - Turning oil resources into reserves
9
© OECD/IEA, 2008
CHAPTER 10
FIELD-BY-FIELD ANALYSIS OF OIL PRODUCTION Is decline accelerating? H
I
G
H
L
I
G
H
T
S
The future rate of decline in output from producing oilfields as they mature is a critical determinant of the amount of new capacity and investment that will be needed globally to meet projected demand. A detailed field-by-field analysis of historical production trends reveals that the size of reserves and physiographic situation (onshore/offshore) are the main factors in explaining the shape of an oilfield’s production profile. The larger the reserves, the lower the peak relative to reserves and the slower the decline once a field has come off plateau. Rates are also lower for onshore than offshore (especially deepwater) fields. Based on data for 580 of the world’s largest fields that have passed their production peak, the observed decline rate — averaged across all fields and weighted by their production over their whole lives — is 5.1%. Decline rates are lowest for the biggest fields: they average 3.4% for super-giant fields, 6.5% for giant fields and 10.4% for large fields. The average rate of observed post-plateau decline, based on our data sub-set of 479 fields, is 5.8%. Observed decline rates vary markedly by region. Post-peak and post-plateau rates are lowest in the Middle East and highest in the North Sea. This reflects, to a large extent, differences in the average size of fields, which in turn is related to the extent to which overall reserves are depleted and their physiographic location. In general, observed decline rates are also higher the younger the field, mainly because these fields tend to be smaller and are more often found offshore. Investment and production policies also affect decline rates, especially in OPEC countries. The average size of the fields analysed is significantly larger than that of all the fields in the world, as our database contains all the super-giant fields and most of the giant fields. The decline rates for the fields not included in our dataset are, on average, likely to be at least as high as for the large fields in our database. On this basis, we estimate that the average production-weighted observed decline rate worldwide is 6.7% for post-peak fields.
Our Reference Scenario projections imply a one percentage-point increase in the global average natural decline rate to over 10% per year by 2030 as all regions experience a drop in average field size and most see a shift in production to offshore fields. This means that total upstream investment in some countries will need to rise, in some cases significantly, just to offset decline.
Chapter 10 - Field-by-field analysis of oil production
221
© OECD/IEA, 2008
The average annual natural, or underlying, decline rate for the world as a whole — stripping out the effects of ongoing and periodic investment — is estimated at 9% for post-peak fields. In other words, the decline in production from existing fields would have been around one-third faster had there been no capital spending on those fields once they had passed their peak.
Understanding production patterns and trends An understanding of oilfield production profiles and the impact of various geological and economic variables on the shape of production curves is critically important to projecting future output from fields already in production or from fields that are yet to be brought into production. A major finding of past Outlooks is that the future rate of production decline from producing fields aggregated across all regions is the single most important determinant of the amount of new capacity that needs to be added and the need to invest in developing new fields (see, in particular, IEA, 2001 and 2003). In other words, future supply is far more sensitive to decline rates than to the rate of growth in oil demand.1 Most investment over the projection period will actually be needed to offset the loss of capacity from existing fields as they mature, oilfield pressure declines and — in the absence of new investment — well flow-rates fall (see Chapter 13 for a detailed analysis of investment trends). For these reasons, we have undertaken a detailed study of historical oilfield production trends, using extensive field-by-field data, with a view to achieving a better understanding of the drivers of decline rates, how they could develop in the future and what that will mean for investment. The results of this study were used to model future production levels to 2030, the results of which are set out in detail in Chapter 11. The field-by-field study involved building a large database containing the full production history and a range of technical parameters for around 800 of the largest individual oilfields in the world. The database includes, to the best of our knowledge, all the super-giant fields, containing initial proven and probable (2P) reserves2 exceeding 5 billion barrels; virtually all giant fields, containing more than 500 million barrels, in production today; and the bulk of the world’s large fields, containing at least 100 million barrels. Together, these fields account for close to three-quarters of all the initial reserves of all the fields ever discovered worldwide and more than two-thirds of all the crude oil produced globally in 2007 (see Box 10.1 for more details). We believe that this is the most comprehensive study of field-by-field oil production patterns and trends ever made public.3 We intend to extend and refine this work in the future.
1. See Chapter 11 for the results of an analysis of the sensitivity of oil production and prices to changes in decline rates. 2. All reserves figures cited in this chapter are 2P unless otherwise stated. 3. IHS/CERA prepared a study on oilfield decline rates based on data for 811 fields for private clients in 2007 (CERA, 2007). 4. Azadegan and Ferdows/Mound/Zagheh in Iran, Kashagan in Kazakhstan, and Tupi in Brazil are not yet producing.
222
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
The majority of the fields that have ever been found worldwide have already been brought into production. The share is highest for the biggest fields, both in terms of their number and their overall reserves, mainly because all but a handful of them were discovered several decades ago. Out of an estimated 58 super-giant fields that have been found, all but four4 are already in production; out of close to 400 giant fields that we have identified, around 80 are either being developed, waiting to be developed
or are temporarily shut in.5 The combined initial 2P reserves of all super-giant and giant fields amount to 1 306 billion barrels, of which an estimated 697 billion barrels remain (equal to about half of the world’s remaining reserves of conventional oil — see Chapter 9). In total, an estimated 79% of the world’s remaining conventional oil reserves are contained in fields that are already being exploited. Thus, the outlook for production at these fields is critical to world oil supply in the short to medium term. S P O T L I G H T
What do rising decline rates mean for oil production and investment? The production profile of an oilfield reflects a number of different factors, including the techniques used to extract its reserves, the field-development programme, reservoir management practices, geology, national production policies, field-maintenance programmes and external factors, such as strikes and military geopolitical conflicts. The analysis presented below points to a long-term trend towards higher faster decline rates once oilfields have reached their peak, as a result of a shift in the pattern of fields that will be brought on stream in the future. In particular, a growing share of production is expected to come from smaller fields and, in the medium term, from fields located offshore, which tend to decline much more quickly than big, onshore fields because of the way they are developed. This is, in turn, largely a function of technical and economic factors rather than geology. Faster decline rates go hand-in-hand with higher peak-production levels relative to reserves.
10
5. These figures do not include some new fields, including offshore Brazil, as they are yet to be properly appraised.
Chapter 10 - Field-by-field analysis of oil production
223
© OECD/IEA, 2008
Rising decline rates have important implications for development costs and investment needs. In general, the smaller a field, the more expensive it is to develop (and operate), expressed in dollars per barrel per day of capacity and, especially, per barrel of oil produced. Similarly, costs are typically significantly higher for offshore fields, particularly in deep water. Over the projection period, all regions continue to experience a drop in average field size and most see a shift in production to offshore fields as the biggest fields, which are usually found and developed first, decline. This means that total upstream investment in some countries will need to rise, in some cases significantly, just to offset decline (even though total world investment needs are expected to drop, as the share of the lowest-cost producing regions in total production increases). The biggest increases are needed in OPEC countries. It is far from certain that all the required investment will be forthcoming, given the size of these investments and potential barriers (see Chapter 13) — because of so-called “above-ground” factors and not because of geology. These factors are discussed in detail in the next four chapters.
Current estimates of reserves are derived from the latest estimates of the oil initially in place and the share that can be economically recovered. Yet that share, as well as the rate at which that share can be produced, are far from certain, as the precise behaviour of a given field or reservoir is exceedingly difficult to predict. Moreover, the deployment of new production technologies (including secondary and enhanced recovery techniques) can push up ultimate recovery rates and production levels. Careful analysis of these factors is vital to predicting future recovery and decline rates and, therefore, production.
Box 10.1
The IEA field-by-field oil production database
The IEA has compiled a database containing the full crude oil production history and a range of key technical parameters for a total of 798 oilfields worldwide. To the best of our knowledge, the database includes all of the world’s 54 super-giant fields that have ever produced, as well as the bulk of the giant producing fields (263 out of a total of around 320).6 Of the remaining 481 fields, 285 are large fields, representing at least half of all the fields in this category in the world and most of the largest ones. The rest of the fields are small fields, containing between 50 and 100 million barrels. The choice of which large and small fields to include in the database was partly driven by data availability. Nonetheless, we believe that the dataset for large fields is reasonably representative of the actual geographic distribution of all such fields in production worldwide today. For all producing fields, data was compiled on annual oil production over the full life of the field, initial and remaining proven and probable (2P) recoverable reserves, the volume of oil initially in place, lithology (the geological formation of the field), physiographic location (onshore, offshore and shallow/deep water) and the discovery date. For some fields, additional information on reservoir porosity and thickness, as well as the deployment of improved recovery techniques, was also obtained.
6. The precise number of fields in these categories may differ among data sources, because of differences in the way specific fields are delineated and data discrepancies.
224
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
The field-by-field data were compiled from a range of different sources. The primary source of data on production and reserves was IHS, without whose assistance we would not have been able to carry out this work. Deloitte & Touche Petroleum Services also provided data on a number of fields. The US Geological Survey and the US Energy Information Administration supplied data for some US fields. Other sources include official statistics, published by the governments of oil-producing countries, international and national oil companies, oil services companies and consulting firms. These organisations assisted us in validating and checking the consistency and veracity of the data, as well as in verifying the results of our analysis. We gratefully acknowledge their contribution.
The importance of size There are currently about 70 000 oilfields in production worldwide. The bulk of crude oil production comes from a small number of very prolific fields, mostly supergiants and giants.7 Output at the world’s ten largest producing oilfields totalled just over 14 million barrels per day (mb/d) in 2007 (Table 10.1), contributing 20% of world conventional production. The 20 largest fields produced 19.2 mb/d, or over a quarter of world production. One field alone — Ghawar in Saudi Arabia — produced 5.1 mb/d, equal to 7% of world conventional oil production (see the next section).
Table 10.1
The world’s 20 biggest oilfields by production
Field
Country
Location
Year of discovery
Peak annual production
2007 production
Year
kb/d
Ghawar
Saudi Arabia
Onshore
1948
1980
5 588
Cantarell
Mexico
Offshore
1977
2003
2 054
1 675
Safaniyah
Saudi Arabia
On/off
1951
1998
2 128
1 408
Rumaila N & S
Iraq
Onshore
1953
1979
1 493
1 250
Greater Burgan
Kuwait
Onshore
1938
1972
2 415
1 170
Samotlor
Russia
Onshore
1960
1980
3 435
903
kb/d 5 100
Ahwaz
Iran
Onshore
1958
1977
1 082
770
Zakum
Abu Dhabi (UAE)
Offshore
1964
1998
795
674
Azeri-Chirag-Guneshli
Azerbaijan
Offshore
1985
2007
658
658
Priobskoye
Russia
Onshore
1982
2007
652
Top 10 total
10
652 14 260
Bu Hasa
Abu Dhabi (UAE)
Onshore
1962
1973
794
550
Marun
Iran
Onshore
1964
1976
1 345
510
Raudhatain
Kuwait
Onshore
1955
2007
501
501
Gachsaran
Iran
Onshore
1928
1974
921
500
Qatif
Saudi Arabia
On/Off
1945
2006
500
500
Shaybah
Saudi Arabia
Onshore
1968
2003
520
500
Saertu (Daqing)
China
Onshore
1960
1993
633
470
Samotlor (Main)
Russia
Onshore
1961
1980
3 027
464
Fedorovo-Surguts
Russia
Onshore
1962
1983
1 022
458
Zuluf
Saudi Arabia
Offshore
1965
1981
677
Top 20 total
450 19 163
7. In this report, a super-giant is defined as a field with initial 2P reserves of at least 5 billion barrels. A giant is defined as a field with initial reserves of 500 million barrels to 5 billion barrels. A large field contains more than 100 million barrels.
Chapter 10 - Field-by-field analysis of oil production
225
© OECD/IEA, 2008
Sources: IHS, Deloitte & Touche and USGS databases; other industry sources; IEA estimates and analysis.
Four of the other fields are also in Saudi Arabia and eight others in other Middle Eastern countries (Iran, Iraq, Kuwait and the United Arab Emirates). Output in 2007 at 16 of the 20 largest fields was below their historic peaks. Production has fallen most in percentage terms at Samotlor in Russia. All of the 20 largest producing fields are super-giants, of which Ghawar, with 140 billion barrels of initial reserves, is by far the largest. Most of the world’s largest fields — by production and reserves — have been in production for many years, in some cases for several decades. The last of the top 20 producing fields to be discovered was Azeri-Chirag-Guneshli in 1985. Priobskoye in Russia was found in 1982, Canterell in Mexico in 1977, and all the others between 1928 and 1968. In 2007, five fields produced more than 1 mb/d and another eight more than 500 thousand barrels per day (kb/d). They make up one-quarter of world crude oil production. Around 110 fields in total produce more than 100 kb/d each. Collectively, they account for just over 50% of world production. A very large number of small fields, each producing less than 100 kb/d at present — approximately 70 000 in total — produce just under half of world production.
Table 10.2
World crude oil production by output and age of field Number of fields
> 1 mb/d 500 kb/d — 1 mb/d All fields
Year of first production Pre-1970s
1970s
1980s
Production, 2007
1990s
2000s
mb/d
%
5
4
1
-
-
-
10.6
15
11
8
-
1
2
-
6.7
10
70 000
n.a.
n.a.
n.a.
n.a.
n.a.
70.2
100
The world’s oil supplies remain very dependent on output from big, old fields. Despite the fact that many of them have been in production for decades, output from super-giant and giant fields (holding more than 500 million barrels of initial reserves) has actually grown significantly over the past two decades. The share in world production of all the super-giant fields and those giant fields included in our database rose from 56% in 1985 to 60% in 2007. Surprisingly, fields that came into production before the 1970s still make the largest contribution, amounting to just over 24 mb/d in 2007 — equal to 35% of the world total (Figure 10.1). Indeed, output from these fields has gradually risen since the mid-1980s (it fell sharply in the early 1980s, mainly because of OPEC policies). Only five super-giant or giant fields began producing in the current decade — Ourhoud in Algeria, Grane in Norway, Girassol in Angola, Jubarte in Brazil and Xifeng in the Gansu province of China — and made up a mere 2% of total output from such fields and little more than 1% of world output in 2007. 226
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
Sources: IHS, Deloitte & Touche and USGS databases; other industry sources; IEA estimates and analysis.
mb/d
Figure 10.1
World crude oil production from super-giant and giant fields by field vintage
50
2000s 1990s
40
1980s 1970s
30
Pre-1970s 20 10 0 1970
1980
1990
2000
2007
Note: For fields covered by IEA field-by-field oil production database (which includes all the world’s supergiant fields and most giant fields). Fields are classified according to the year of first production. Sources: IHS and Deloitte & Touche databases; other industry sources; IEA estimates and analysis.
10
Regional differences Big oilfields are unevenly distributed across the world; their share in overall production and their average size vary markedly across regions. The Middle East is characterised by a large number of super-giant and giant fields. With 9 billion barrels of initial reserves, their average size is the highest of any region (Table 10.3).
Geographical distribution of the world’s super-giant and giant oilfields (number) Super-giants and giants
Of which offshore
Average size of total (billion barrels)
OECD North America
46
11
2.0
OECD Europe
23
23
1.4
OECD Pacific E. Europe/Eurasia
2
2
1.1
62
5
3.1
Asia
20
5
2.1
Middle East
83
25
8.9
Africa
41
12
1.7
Latin America
40
6
3.4
317
89
4.2
Total
Note: For fields covered by IEA field-by-field oil production database. See footnote 7 for definitions of supergiant and giant fields. Sources: IHS, Deloitte & Touche and USGS databases; other industry sources; IEA estimates and analysis.
Chapter 10 - Field-by-field analysis of oil production
227
© OECD/IEA, 2008
Table 10.3
The region holds a quarter of all super-giant and giant fields. Around three-quarters of the world’s super-giant and giant fields are located onshore (including fields that straddle land and sea). The share is highest in the Middle East, Asia and the former Soviet Union. In Europe, all big fields are located offshore. Super-giant and giant fields account for the largest share of production in the Middle East, Russia and the Caspian region (Eastern Europe/Eurasia) and Latin America (Figure 10.2). Their share is lowest in Asia, Europe and the Pacific region. Although North America accounts for just over a quarter of all the crude oil ever produced in the world and 13% of current output, there are little more than 50 super-giant and giant fields in that region — a far smaller number relative to output than in any other region.8
Figure 10.2
Crude oil production by region and size of field, 2007 Super-giant
OECD Pacific
Giant
OECD Europe
Other Latin America Asia Africa OECD North America E. Europe/Eurasia Middle East 0
5
10
15
20
25 mb/d
Note: For fields covered by IEA field-by-field oil production database. See footnote 3 for definitions of supergiant and giant fields. Sources: IHS and Deloitte & Touche databases; other industry sources; IEA estimates and analysis.
8. The highly fragmented ownership of North American fields means that many large formations are broken down statistically into separate fields, resulting in a smaller number of giants.
228
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
There are also big differences, according to their geographic location and their size, in the extent to which reserves have been depleted. Among all fields in production today, the depletion factor — the share of initial reserves that has already been produced — is marginally higher for the super-giant and giant fields. Worldwide, those fields are on average 48% depleted (weighted by total production), compared with 47% for other fields included in our study (Table 10.4). Depletion factors are highest for North America, where most fields have been in production for decades, and Europe, where small fields dominate production. They are lowest in the Middle East.
Table 10.4
Average depletion factor of producing fields* by size, 2007 Super-giants and giants
Others
All fields
OECD North America
78%
83%
81%
OECD Europe
77%
71%
73%
Middle East
37%
14%
32%
Africa
61%
44%
50%
Total
48%
47%
48%
* Based on the full IEA dataset of 798 fields. Note: The depletion factor is cumulative production divided by initial 2P reserves. Sources: IHS, Deloitte & Touche and USGS databases; other industry sources; IEA estimates and analysis.
Oilfield production profiles and characteristics Every oilfield follows a unique production profile, according to the natural characteristics of the reservoirs within it, the manner in which it is developed and production-management policies. Typically, an oilfield goes through a build-up phase, during which production rises as newly drilled wells are brought into production, a period of plateau production, during which output typically is broadly flat as new wells are brought on stream offsetting declines at the oldest producing wells, and a decline phase, during which production gradually falls with reservoir pressure. In practice, oilfields rarely follow a smooth, predictable production path. Commercial and policy considerations affect how a field is developed. And reservoirs behave in different ways at different stages of depletion for geological and technical reasons. In addition, production rates can fluctuate sharply as new phases of a field’s development are launched, often to combat the “natural” or underlying decline in output. In general, with larger fields, the build-up period is long and development is pursued in phases. Some fields can build up over several decades: the Zakum field in the United Arab Emirates, for example, began producing in 1967 and hit record output of 790 kb/d only in 2002 — a level that is expected to be exceeded in the future. Periodic maintenance programmes (scheduled and unplanned) and deliberate shut-ins for policy reasons (for example, to comply with national production quotas) can also upset the underlying trend in production.
10
Distinguishing the impact of the inherent technical characteristics of specific types of fields from the effect of how those fields are developed and managed over time is crucial to understanding historic production trends and assessing the long-term prospects for output — both for fields that are already producing and those that are yet to be developed. For this reason, we have identified standard production profiles for different types of oilfields and assessed how those profiles differ according to a number of technical variables. This analysis is based on a sample of 725 fields from our oilfield database, with an average production history of just under 22 years and total initial reserves of 1 358 billion barrels (Table 10.5) The oldest field, Balahani-SabunchiChapter 10 - Field-by-field analysis of oil production
229
© OECD/IEA, 2008
Standard production profiles
Ramani in Azerbaijan, started producing in 1871. The full dataset could not be used, because of a lack of data on certain technical parameters. Nonetheless, almost all super-giant and giant fields are included, together with most large fields. Table 10.5 By location Onshore* Offshore shelf Offshore deepwater By lithology Carbonate Sandstone Chalk Unknown Total
Initial reserves of oilfield dataset for production profiling Number of fields
2P reserves (billion barrels)
400 294 31
1 120 213 25
145 473 12 95 725
716 613 6 23 1 358
* Includes fields partially offshore. Sources: IHS, Deloitte & Touche and USGS databases; other industry sources; IEA estimates and analysis.
The notional average productive life of each category of field (on the assumption that cumulative production equals total initial reserves when the field is abandoned) differs markedly, from 27 years for deepwater fields to 110 years for fields holding more than 1.5 billion barrels. In practice, however, the tail of production at mature fields is strongly influenced by the prevailing economic conditions. The water cut (the share of water contained in the mixture of hydrocarbons and water that flows from the well) tends to rise towards the end of a field’s life, pushing up processing costs. For as long as total operating costs are below the market value of the oil recovered, production can be sustained at relatively low levels for a long time. Since the behaviour of heavily depleted fields varies, the estimates presented here should be considered indicative only. 230
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
The analysis revealed that the size of reserves and the physiographic situation (onshore/ offshore) are the most important variables in explaining the shape of the production profiles (Box 10.2). Lithology — essentially whether the field is carbonate or sandstone — does not appear to influence to a significant degree the shape of the production profile, everything else being equal. The results show that small fields reach their peak sooner, produce a higher share of initial reserves by peak and decline more rapidly than large fields (Table 10.6 and Figure 10.3). It takes about twice as long for big fields to get to peak. Everything else being equal, peak production relative to reserves at offshore (shelf) fields is higher than at onshore fields, reflecting the need for developers to recover more quickly the higher costs usually associated with offshore fields. Deepwater fields, although usually big, behave in a similar way to small offshore fields with peak production reached after five years. On average, 7% of reserves are produced in that year, with cumulative production reaching 22% of reserves. The production curve for deepwater fields is, thus, highly skewed to the left, with less than a quarter of reserves produced in the relatively brief pre-peak period. Generally, offshore fields tend to have fewer wells but more highly productive horizontal wells. Spacing between wells is also much wider for offshore fields, because of the higher cost of drilling wells.
Box 10.2
Oilfield production-profiling methodology
The analysis of oilfield production profiles seeks to explain the shape of standardised profiles according to a number of technical variables, namely: Total crude oil reserves. Physiographic situation (onshore, shallow offshore and deep water). Lithology (carbonate, sandstone and chalk). Permeability, thickness and porosity of the reservoir and the API gravity of the liquids present (for fields holding initial reserves of at least 1 billion barrels). These parameters were amalgamated into a single indicator of the transmissibility of the fluid in the reservoir rock, by multiplying permeability and thickness with each other and dividing the result by gravity. The analysis revealed that the size of reserves and the physiographic situation were the most important variables in determining the shape of the production profile. Therefore, normalised production curves (plotting annual production against cumulative production, both expressed as the share of initial 2P reserves) were first estimated according to the size of reserves for three categories: over 1.5 billion barrels; between 500 million and 1.5 billion barrels; and less than 500 million barrels. Using statistical techniques, the degree of influence of the other technical variables in explaining the shape of the normalised production curves was then estimated.9 The results were used to produce mean normalised curves for each size category of field according to different variants of the technical variables. The mean curves were then extended to show how each category of field would be expected to behave through to full depletion of reserves. This was done by assuming an exponential rate of change. Table 10.6
10
Production characteristics of sample oilfield dataset for production profiling % of initial Cumulative % of Number of years Estimated average reserves produced initial reserves of production total number in the peak year produced in the at plateau* of production peak year years** 3.9
21
7
Shelf, 1.5 Gb
1.7
15
13
110
Deepwater
7.0
22
5
27
* Defined as the period during which production is more than 85% of that in the peak year. ** Over the full life of the field, assuming that cumulative production strictly equals initial reserves. Sources: IHS, Deloitte & Touche and USGS databases; other industry sources; IEA estimates and analysis. 9. A detailed explanation of the procedures used can be found at www.worldenergyoutlook.org.
Chapter 10 - Field-by-field analysis of oil production
231
© OECD/IEA, 2008
Onshore, < 500 Mb
Annual production as share of initial 2P reserves
Figure 10.3
Standard oilfield-production profiles by category of field
10%
Onshore 2.8 Gb - total
1.5% 1.0% 0.5% 0% 0%
10%
20%
30% 40% 50% 60% 70% 80% 90% 100% Cumulative production as share of initial 2P reserves
Note: the thick lines are derived from observed data while the thin lines show the trajectory assuming full depletion of the field.
Changes in production profiles over time The standard production profiles for the six categories of field analysed above (Figure 10.3) are derived from production data for fields that came into production at different times over several decades. Three-quarters of the fields with reserves of more than 1.5 billion barrels in this dataset started to produce in the 1970s or earlier. More than 60% of medium-sized fields, both onshore and offshore, also started producing in the same period. The smallest fields were generally developed later: half of small onshore fields and more than three-quarters of small offshore fields started to produce in the last 15 years.
10
Analysis of production profiles by vintage shows that fields developed in recent years tend to build up more quickly to a higher plateau (relative to reserves), maintained over a shorter period of time, than fields developed before the 1990s. For example, Hassi Berkine Sud, a medium-sized Algerian field brought into production in 1998, recently reached plateau production equal to around 6% of reserves compared with a typical peak of little more than 2% for all onshore fields of similar size in our dataset (Figure 10.5). This finding is not particularly surprising: advances in production technology have made it financially attractive to introduce improved and enhanced recovery techniques earlier in a field’s life. In addition, the pressure from investors on private companies to minimise the payback period and so maximise the net present value of future cash-flow, has increased in recent years.
Approach and definitions While each phase of an oilfield’s life is important, the rate at which production from oilfields declines once it has reached peak production is critical to determining the Chapter 10 - Field-by-field analysis of oil production
233
© OECD/IEA, 2008
Measuring observed production decline rates
need for additional capacity, either through further development work at existing fields or by bringing new fields into production. At a global level, the faster the rate of decline, the greater the need for additional capacity for a given level of demand. The actual rate of decline, how it has changed and how it could evolve in the years to come have assumed enormous importance in the current debate about the medium- and long-term prospects for oil supply. For this reason, we have carried out an in-depth, field-by-field analysis of decline rates in order to improve understanding about the topic and gain insights into future production trends.
Annual production as share of initial 2P reserves
Figure 10.5
Selected production profiles of recently developed mediumsized onshore fields compared with the standardised profile
7%
Tyanskoye - 1995
6%
Castilla - 2000
5%
Hassi Berkine Sud - 1998
4%
Medium onshore fields standard profile
3% 2% 1% 0% 0%
10%
20%
30%
40% 50% 60% 70% 80% 90% 100% Cumulative production as share of initial 2P reserves
Of a total of 798 producing fields in our field-by-field database, we prepared a dataset of 651 fields with initial reserves of at least 50 million barrels in order to carry out our analysis of decline rates. Of this set, 580 fields were found to have passed peak production (Table 10.7). In other words, for each of these fields, production over the latest year of production is below the maximum level ever achieved in any one year. These fields produced a total of 40.5 mb/d in 2007, or 58% of world crude oil production. Their initial reserves total 1 241 billion barrels, equal to 52% of the world total. Of the post-peak fields, a total of 479 were found to be in the post-plateau phase. Of these, 362 fields are in decline phase 3 (with annual production at less than 234
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
Discussions about decline rates are often confused by a failure to make clear what is meant by the term and exactly how they are calculated. It is important to understand that, at any given moment, some oilfields will be ramping up to peak production, others will be at peak or plateau, and others will be in decline. Averaging rates across a group of fields does not, therefore, reveal by itself any clear information on the decline rate of fields at different stages of their production life. Only a field-by-field analysis of production trends can shed light on this. In this section, we use our field-by-field database to quantify decline rates in detail and analyse long-term trends. The subsequent section looks at how decline rates could evolve over the projection period. The precise definitions and methods used to measure decline rates are described in Box 10.3.
Box 10.3
How do we define and calculate decline rates?
Peak production is the highest level of production recorded over a single year at a given field. An oil field is in decline when aggregate production in the latest year (2007 for most fields in the dataset used for this analysis) is below production in the peak year, even if the field attains other lower peaks in the interim. Plateau production is when annual production is more than 85% of peak production. A field is in the post-plateau phase when it has fallen below plateau. For the purposes of our decline rates calculations, only fields with production in the last year of production that is below the level of the first year of post-plateau production are included. The full period of production decline after peak is broken down into three distinct phases for the purposes of measurement: decline phase 1 is the period from peak annual production to the last year of plateau production; decline phase 2 is from the first year at which production falls below plateau through to the last year in which production is above 50% of peak; and decline phase 3 is the period after which production is consistently below 50% of peak. The observed decline rate is the cumulative average annual rate of change in observed production between two given years (for example, between peak production and the latest year). The natural decline rate, sometimes called the underlying decline rate, is the notional rate of decline in production between two given years had there been no investment beyond that associated with the initial development of the field. The methodology used to estimate this rate by region is described in Figure 10.9 below.
10
Unless otherwise mentioned, all the decline rates referred to in this chapter are production-weighted. In other words, the average for a particular group of fields (by type or region) takes into account the level of production of each field in the total. Cumulative production over the full life of the field was used to weight decline rates for fields currently in production and in the post-peak phase.
half that at peak). A larger share of super-giant fields are in the post-plateau phase, as most of them have been in production for several decades. Nonetheless, the world’s biggest field by far — Ghawar — is not among the post-plateau fields, as production in 2007 was still less than 15% below the peak of 5.6 mb/d reached in 1980 (Box 10.4). Chapter 10 - Field-by-field analysis of oil production
235
© OECD/IEA, 2008
Generally, historical observed decline rates were calculated using the full production history of each field. Of course, these vary greatly in length: the oldest field in our dataset has been producing for 137 years and the youngest for two years (the minimum period for which calculating a decline rate is possible). Solely for the purposes of measuring long-term trends, decline rates were also calculated on a year-by-year basis, with the decline rates for each field weighted by the actual production in the given year.
Table 10.7
Number of oilfields in dataset for decline rate calculations Super-giant
Giant
Other
All fields
Onshore*
43
185
159
387
Offshore shelf
11
61
147
219
0
17
28
45
By location
Offshore deepwater By lithology Carbonate
32
69
59
160
Sandstone
22
189
268
479
0
5
7
12
40
97
48
185
33
41
8
82
Chalk By grouping OPEC Middle East Other
7
56
40
103
Non-OPEC
14
166
286
466
By region OECD
3
68
150
221
3
43
56
102
Europe
0
23
89
112
Pacific
0
2
5
7
Non-OECD
51
195
184
430
10
52
14
76
1
19
73
93 101
North America
E. Europe/Eurasia Asia Middle East
33
50
18
Africa
1
40
53
94
Latin America
6
34
26
66
54
263
334
651
Total
* Includes fields partially offshore. The dataset includes all post-peak fields in our database. Sources: IHS, Deloitte & Touche and USGS databases; other industry sources; IEA estimates and analysis.
The observed post-peak decline rate averaged across all fields on a production-weighted basis is 5.1% using raw data. That is equal to 3.6 mb/d per year, based on the 2007 level of global crude oil production. As the standard production profiles suggest, decline rates are lowest for the biggest fields: they average 3.4% for super-giant fields, 6.5% for giant fields and 10.4% for large fields (Table 10.8). Rates are also lowest for onshore fields and highest for deepwater offshore fields, reflecting the different ways in which they are developed (as described in the previous section). On average worldwide, production has declined yearly by 4.3% at onshore fields and 7.3% at offshore fields, with deepwater fields declining by 13.3%. Sandstone fields have declined significantly faster than carbonate fields, 6.3% versus 3.4%, but this largely reflects the fact that the latter tend to be much larger and are more often located in the Middle East (where decline rates have been tempered by historically conservative production policies); 236
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
Results of the analysis
Box 10.4
The Ghawar field: the super-giant among super-giants
The Ghawar field — the world’s largest — was discovered in 1948 and started producing in 1951. The area of the field — more accurately described as a collection of oil-bearing formations — is partitioned into six geographical areas, from north to south: Ain Dar, Shedgum, Farzan, Hawiyah, Uthmaniyah and Haradh. Oil is produced from the Jurassic formations, namely Arab, Dhruma and Hanifa, while gas and condensate are extracted from the deeper and older reservoirs, in the Khuff, Unayzah and Jawf formations. Ghawar crude, which has an average gravity of around 34° API and sulphur content of 1.8%, accounts for most of the Saudi Arab Light export blend.
Reservoir pressure is maintained through the use of peripheral water flooding, whereby seawater is injected into the reservoir in the oil layer just above the tar mat that separates the oil layer from the aquifer. The water pushes the oil inwards and upwards, towards the producing wells. This secondary recovery technique, first used at Ghawar in 1965, typically results in lower flow rates than the more commonly used pattern water flooding, but tends to result in a higher recovery rate and allows for plateau production to be maintained for longer (see IEA, 2005 for more details). Nonetheless, as the field has matured, maintaining reservoir pressure and sustaining production has become more difficult and costs have risen. The water cut — the share of water in the liquids extracted — increased sharply in the 1990s, reaching 37% in 2000. However, recent development work has succeeded in reducing the water cut to 27% at present, according to Saudi Aramco, the field operator. Ghawar has been developed in distinct stages, which have progressively raised the field’s capacity and kept the field at plateau. The most recent project, involving the Haradh area in the southern part of the field, was completed in 2006, tripling capacity there to about 900 kb/d. This has helped to offset natural declines in other parts of the field. The overall capacity of Ghawar is sustained by infill drilling and well work-overs to maintain flow pressure in various parts of the field. Reports suggest that enhanced oil recovery techniques are being used to boost capacity in the mature zones of the Shedgum and Uthmaniyah areas, where extensive drilling programmes have recently been undertaken (Sanford Bernstein, 2007).
Chapter 10 - Field-by-field analysis of oil production
237
10
© OECD/IEA, 2008
Ghawar is a large anticline structure, 280 km long by 25 km wide, with about 50 metres of net oil pay. Initial oil in place is 250 billion barrels, of which initial recoverable reserves are estimated at 140 billion barrels (implying an expected ultimate recovery rate of 56%). Cumulative production reached 66 billion barrels in 2007, so remaining reserves are about 74 billion barrels. Ghawar produced 5.1 mb/d of crude oil in 2007, down from a peak of 5.6 mb/d in 1980 (when the field’s capacity was fully utilised in response to the loss of Iranian production following the revolution) and a recent peak of 5.3 mb/d in 1997. The observed post-peak decline rate is, thus, a mere 0.3% per year. Ghawar is still at the plateau phase of production on our definition (Box 10.3).
lithology does not appear to be a major determinant of decline rates (see the previous section). Regardless of location or lithology, decline rates are, in most cases, lower the bigger the field. Similarly, the rates of post-plateau decline are usually slightly higher than the rate of post-peak decline, confirming the results of the production-profiling analysis (Figure 10.6).10 Worldwide, the average rate of observed post-plateau decline is 5.8% for the 479 fields that are in the post-plateau decline phase out of the 580 postpeak fields and the total of 798 fields in our database.11
Table 10.8
Production-weighted average observed decline rates by size and type of field Post-peak Supergiant
Giant
Onshore
3.4%
Offshore
3.4%
Shelf
Post-plateau
Large
Total
Supergiant
Giant
5.6%
8.8%
4.3%
4.9%
5.5%
9.4%
5.3%
8.6%
11.6%
7.3%
1.2%
9.0%
11.7%
7.2%
3.4%
7.7%
11.2%
6.6%
1.2%
8.6%
12.2%
6.7%
–
13.1%
14.2%
13.3%
–
10.8%
12.6%
11.2%
Carbonate
2.3%
6.6%
8.9%
3.4%
2.7%
6.9%
9.3%
4.3%
Sandstone
4.8%
6.5%
10.9%
6.3%
5.5%
6.5%
11.1%
6.6%
World
3.4%
6.5%
10.4%
5.1%
4.3%
6.6%
10.7%
5.8%
Deepwater
Large
Total
Sources: IHS, Deloitte & Touche and USGS databases; other industry sources; IEA estimates and analysis.
10. The average post-peak decline rates calculated using the standard production profiles for different categories of field by size are, unsurprisingly, of a similar magnitude to the rates calculated using raw data that are presented in this section. 11. The observed decline rates shown in this chapter differ from those for non-OPEC countries described in the IEA’s most recent Medium-Term Oil Market Report, published in July (IEA, 2008). This is because of methodological differences, associated with the different time horizons of the two reports. The MTOMR estimates decline rates on a year-by-year basis, rather than over the full production life of the field. The MTOMR also adjusts raw field-by-field production data for temporary reductions in output caused by unexpected events such as weather-related outages, strikes and security-related disruptions. Nonetheless, the results are similar.
238
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
Decline rates also vary markedly by region. Rates are lowest in the Middle East and highest in the North Sea (Table 10.9). This reflects, to a large extent, differences in the average size of fields, which in turn is related to the extent to which reserves are depleted and their location onshore or offshore. North Sea fields tend to be much smaller than Middle East fields, while almost all significant North Sea fields are located offshore. North Sea fields (which make up all the European fields in our dataset) have declined on average by 11.5% per year since peak and 13.3% since plateau. The relatively low decline rates of Middle East fields, which have averaged less than 3% per year, is also explained by the
disruptions to the standard production profile caused by short-term productionmanagement policies (notably in support of OPEC targets) and geopolitical conflicts. The dominance of Middle East countries and the heavy weight of super-giant fields in the population of OPEC fields contribute to the much lower average post-peak decline rates for OPEC compared with non-OPEC countries.
Figure 10.6
Production-weighted average post-peak and post-plateau observed decline rates by field size
12%
Post-peak
10%
Post-plateau
8% 6% 4% 2% 0% Giant
Super-giant
Large
Total
10
Sources: IHS, Deloitte & Touche and USGS databases; other industry sources; IEA estimates and analysis.
Production-weighted average annual observed decline rates by region Post-peak Supergiant
OPEC
Giant
Large
Post-plateau Total
Supergiant
Giant
Large
Total
2.3%
5.4%
9.1%
3.1%
2.9%
4.8%
8.3%
3.6%
Middle East
2.2%
6.3%
4.4%
2.6%
2.8%
6.5%
6.4%
3.4% 4.3%
Other
4.8%
5.0%
10.2%
5.2%
3.8%
4.1%
8.8%
Non-OPEC
5.7%
6.9%
10.5%
7.1%
6.0%
7.4%
10.9%
7.4%
OECD North America
6.4%
5.4%
12.1%
6.5%
4.5%
6.0%
12.3%
6.0%
OECD Europe
–
10.0%
13.5%
11.5%
–
13.1%
15.5%
13.3%
OECD Pacific
–
11.1%
13.2%
11.6%
–
10.4%
12.6%
11.1%
E. Europe/Eurasia
5.1%
5.0%
12.1%
5.1%
5.3%
5.1%
12.4%
5.3%
Asia
2.1%
8.3%
6.6%
6.1%
2.5%
5.7%
6.7%
5.2%
Middle East
2.2%
6.5%
7.4%
2.7%
2.8%
7.0%
9.8%
3.7%
Africa
1.5%
5.2%
8.8%
5.1%
1.2%
5.2%
9.3%
5.0%
Latin America
8.4%
5.2%
6.9%
6.0%
9.5%
5.3%
6.8%
6.1%
World
3.4%
6.5%
10.4%
5.1%
4.3%
6.6%
10.7%
5.8%
Sources: IHS, Deloitte & Touche and USGS databases; other industry sources; IEA estimates and analysis.
Chapter 10 - Field-by-field analysis of oil production
239
© OECD/IEA, 2008
Table 10.9
The impact of field age and maturity In general, observed decline rates are higher the younger the field. For example, the average decline rate for all the non-OPEC fields in our dataset that have come on stream since the start of the current decade is 14.5%, compared with 11.6% for post-peak fields that started producing in the 1990s and only 5.9% for fields that started producing no later than 1969 (Table 10.10). This pattern applies to both OPEC and non-OPEC fields (Figure 10.7). The very low average decline rate for the pre-1970s vintage of OPEC fields — less than 3% — is influenced strongly by the very low observed post-peak decline rate of the Ghawar field (0.3%). The fall in the decline rate for OPEC fields that came into production since 2000 is explained by the fact that most of them, while past their initial peak, are still at plateau. Table 10.10
Production-weighted average post-peak observed decline rates by vintage* Pre-1970s
1970s
1980s
1990s
2000s
Total
OPEC
2.8%
3.5%
4.6%
7.5%
5.0%
3.1%
Non-OPEC
5.9%
6.8%
8.3%
11.6%
14.5%
7.1%
World
3.9%
5.9%
7.9%
10.6%
12.6%
5.1%
* First year of production. Source: IHS, Deloitte & Touche and USGS databases; other industry sources; IEA estimates and analysis.
As explained in the previous section, advances in production technology and changes in commercial practice mean that fields developed today tend to build up more quickly to a higher plateau, maintained over a shorter period of time, than fields developed before the 1990s. The growing importance since the 1970s of offshore fields (which normally reach a higher peak as a share of reserves than onshore fields) also explains this trend. It follows that the post-peak decline rate from such shorter and accelerated production profiles is higher. Figure 10.7
Production-weighted average post-peak observed decline rates by type of producer and year of first production
16%
OPEC
14%
Non-OPEC
12% 10% 8% 6%
2% 0% Pre-1970s 1970s 1980s 1990s 2000 - 2007 Sources: IHS, Deloitte & Touche and USGS databases; other industry sources; IEA estimates and analysis.
240
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
4%
Decline rates show some variation according to the decline period (Box 10.3 for definitions), though the differences are less marked than for other factors including field size. For all post-peak fields still in decline phase 1 (i.e. whose production in the most recent year is still more than 85% of peak), the average decline rate is 1.4% (Table 10.11). The decline rate rises to 3.6% for those fields in decline phase 2 and to 6.7% in decline phase 3. For all three decline periods, rates are always lowest for the super-giant fields, with those still in decline phase 1 registering an average decline rate of only 0.8%. Once again, the large weight of the Ghawar field in this group of fields helps to lower the overall decline rate. The rise in the observed decline rate as a field reaches the end of its life is probably explained by the increase in the rate of decline in pressure.
Table 10.11
Production-weighted average annual observed decline rates by decline phase Decline phase 1 (peak to end of plateau)
Decline phase 2 (plateau to 50% of peak)
Decline phase 3 (50% of peak to latest year)
Total
Super-giant
0.8%
3.0%
4.9%
3.4%
Giant
3.0%
3.7%
7.6%
6.5%
Large
5.5%
7.2%
11.8%
10.4%
World
1.4%
3.6%
6.7%
5.1%
Note: See Box 10.3 for precise definitions of the decline periods. Fields were sorted according to the period in which they currently lie (based on the last year of production data). Of the 580 post-peak fields included in our analysis, 101 were in decline phase 1, 117 in phase 2 and 362 in phase 3. Source: IHS, Deloitte & Touche and USGS databases; other industry sources; IEA estimates and analysis.
10
Trends in observed decline rates
Short-term fluctuations appear to reflect the impact of the production policies of OPEC countries and cyclical changes in investment in existing fields, resulting from policy, geopolitical factors and fluctuations in oil prices and fiscal terms. The very rapid fall in production in 1980 is explained primarily by the collapse in output at several fields in Iran following the Iranian revolution in late 1979, while the invasion of Kuwait, which Chapter 10 - Field-by-field analysis of oil production
241
© OECD/IEA, 2008
For individual fields, observed decline rates would be expected to change over time, because of changes in well productivity, as pressure drops, and in production policy, and because of investment in field re-development. At any given moment, fields will be at different stages of their production life, so the rate of observed decline averaged over all fields in decline inevitably changes over time. We have calculated the yearon-year change in production, averaged across all the post-peak fields in our dataset (all the decline rates shown up to this point correspond to the average over the life of each field). For each year, only those fields that were in decline were included in the calculation. The results show that the production-weighted average rate of fall in production has fluctuated significantly over time, but has been relatively more stable since the 1980s, at around 5% to 12% per year (Figure 10.8).
led to the loss of all of the country’s production for several months, caused the drop in output to accelerate in 1990 and 1991. The acceleration in the year-on-year rate of fall in production in the late 1990s was the result of a downturn in upstream investment, in response to a slump in prices, while the fall in the rate of decline since 2000 (with the exception of 2006, when production constraints in OPEC countries caused average rates of decline to accelerate) appears to have resulted from higher capital spending on re-developing existing fields. Figure 10.8 1970 0% –2%
1975
Year-on-year change in the production-weighted average production from post-peak fields 1980
1985
1990
1995
2000
2005 2007 Year-on-year 3-year moving average
–4% –6% –8% –10% –12% –14% –16% –18% –20% Note: In contrast to the decline rates shown in the preceding tables and figures, which are based on the average decline rate of each field over a period of time determined by its peak or plateau, the rates of change shown here are calculated using simple year-on-year changes in each field’s production. For each year, the dataset is adjusted to include only those fields that were in the post-peak phase and that were in decline in that year. As a result, the sample size diminishes going backwards in time. For example, the number of fields that were post-peak in 1970 was 36 compared with 580 in 2007. The decline rates may therefore be considered less representative of all rates for all the world’s producing fields for the earliest years than for 2007. Sources: IHS, Deloitte & Touche and USGS databases; other industry sources; IEA estimates and analysis.
Deriving an estimate of the average global observed decline rate
It is impossible to know precisely what the observed decline rates are for these other fields. Nonetheless, it is reasonable to assume that the decline rates are, on average, at least as high as these of the large fields in our database. In reality, they are likely 242
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
Our field-by-field analysis of decline rates allows us to obtain a reasonable estimate of the average decline rates for all the fields in the world, weighted by production. All the decline rates presented so far in this chapter are based on field-by-field production data from our database, covering 798 fields. The average size of these fields — predominantly super-giants and giants — is significantly larger than the average size of all the fields in the world. The 580 fields included in our analysis of post-peak decline rates produced 40.5 mb/d of crude oil in 2007 — equal to 58% of world production. Yet these fields make up less than 1% of all the producing fields in the world.
to be somewhat higher, given that we have detected a clear correlation between field size and the observed decline rate. In order to derive an indicative estimate of the overall decline rate for all the world’s oilfields, we have assumed that the average rate for the fields not included in our database is the same as that for the large fields (which averages 10.4% worldwide). This is a somewhat optimistic assumption, as the differences in decline rates between the three categories of fields we assess here suggest that smaller fields are likely to have higher decline rates than large fields. On this basis, we estimate that the average observed decline rate worldwide is 6.7%. Were that rate to be applied to 2007 crude oil production, the annual loss of output would be 4.7 mb/d. The adjusted decline rate is higher in all regions, markedly so in North America, because our sample for that region is dominated by super-giant and giant fields (Table 10.12). Table 10.12
Estimated production-weighted average annual observed post-peak decline rates for all fields worldwide by region Based on 580 field dataset
OECD North America
All fields
6.5%
9.7%
OECD Europe
11.5%
11.9%
OECD Pacific
11.6%
12.6%
E. Europe/Eurasia
5.1%
5.8%
Asia
6.1%
6.7%
Middle East
2.7%
3.4%
Africa
5.1%
6.8%
Latin America
6.0%
6.6%
World
5.1%
6.7%
10
Sources: IHS, Deloitte & Touche and USGS databases; other industry sources; IEA estimates and analysis.
Trends in natural decline rates Observed decline rates are an important indicator of the performance of oilfields across regions and over time, but, by themselves, they do not reveal underlying trends in field production behaviour. This is because observed rates are heavily influenced by on-going and periodic investment in fields already in production, aimed at maintaining well pressure and flow rates, and improving recovery of oil reserves. In reality, few oilfields are left to produce without further field-development work involving significant amounts of capital expenditure once the initial set of wells has been drilled. This further investment can take the form of infill drilling (to target pockets of oil that prove to be inaccessible from existing wells), well work-overs (major maintenance or remedial treatments, often involving the removal and replacement of the well casing), secondary recovery programmes such as water flooding (the injection of water to push the oil towards producing wells) and gas injection and enhanced oil recovery techniques, such as CO2 injection. Such activities can arrest the natural decline in Chapter 10 - Field-by-field analysis of oil production
243
© OECD/IEA, 2008
Estimating historical trends
pressure and production from a field and may even boost output to a significant degree. It is necessary to estimate the underlying, or natural decline rate — the rate at which production at a field would decline in the absence of any investment — in order to ascertain how much capital needs to be deployed to sustain production or limit observed decline to a particular rate. To arrive at the natural decline rate, therefore, one needs to strip out the effect of new investment beyond the initial capital spending involved in bringing the field into production. We have developed a top-down methodology for estimating historical natural decline rates, based on the estimated impact of the actual investment that has gone into existing fields over the past five years (summarised in Figure 10.9). This approach required detailed estimates of the amount of new capacity coming into production each year over 2003-2007, the associated capital expenditure and the unit cost of incremental capacity at existing fields, based on generic (region-by-region) estimates of finding and development costs and reserve life estimates (drawing on the results of our analysis of investment in Chapter 13). The results were calibrated against our estimates of observed decline rates for all fields worldwide. Inevitably, a degree of judgment was involved, in consultation with industry, in estimating these parameters, given data deficiencies. The results must, therefore, be considered as indicative only. Figure 10.9
Methodology for estimating natural decline rates Crude oil production by region Annual production, 2002-2007
Production from existing fields Deduct from each year’s production after 2002 estimated output from post-2002 projects
Notional production from existing fields net of investment Derive investment in existing fields (in 2002) by subtracting investment in new fields from total spending
Derive incremental output from investment in existing fields by multiplying investment by estimated unit capital costs
Subtract incremental output from total production from existing fields
Year-on-year natural decline rates by region
The production-weighted average annual natural decline rate for the world as a whole is estimated at 9.0% — some 2.3 percentage points higher than the observed decline rate for post-peak fields. In other words, the decline in production from existing fields would have been around one-third faster had there been no capital spending on those fields 244
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
Calculate the year-on-year decline in notional production from existing fields net of investment
once they had passed their peak. Natural decline rates are highest for OECD Europe and Pacific (Australia), and Asia. The rate is lowest for the Middle East. The difference between observed and natural decline rates varies in both absolute and proportionate terms across regions (Figure 10.10). Nonetheless, the ratio of observed to natural decline rates, which averages about 1:1.3, is broadly similar across all regions, suggesting that these estimates are reasonably robust. The smallest difference between observed and natural rates is for Europe (North Sea). This appears to reflect the relatively limited remaining scope for infill drilling compared with other parts of the world. Figure 10.10
Indicative natural decline rates by region Observed post-peak decline rate
Middle East E. Europe/Eurasia
Additional decline without new investment
Africa Latin America Asia OECD Europe OECD North America OECD Pacific –20%
–15%
–10%
–5%
10
0%
Other recent studies support the finding that natural decline rates have been rising. For example, the natural decline rates of fields operated by 15 major oil companies rose on average from 10.6% to 13% between 2001 and 2006 (Goldman Sachs, 2007). The higher rate for these companies is to be expected, since a relatively large share of their production comes from OECD regions (notably North America) and West Africa, where decline rates are highest. The increase in the overall rate of natural decline, based on our analysis, is far from negligible: based on world crude oil production of 70.2 mb/d Chapter 10 - Field-by-field analysis of oil production
245
© OECD/IEA, 2008
Discerning a trend from the short time period we used to estimate natural decline rates is risky. Nonetheless, a clear rising trend does emerge from our analysis: the worldwide average natural decline rate (year-on-year) rose from 8.7% in 2003 to 9.7% in 2007 (the average rate is 9% for the period 2003-2007 as a whole). This result is in line with expectations, as over that period, a growing share of crude oil production came from younger, smaller and offshore fields, which have inherently higher decline rates. Smaller and offshore fields typically exhibit higher observed decline rates, because of the more limited potential for infill drilling, as mentioned in the previous section. But natural decline rates would also be expected to be higher too, as these fields tend to be developed in such a way as to maximise and bring forward peak production in order to improve cash-flow and amortise the large up-front investment as quickly as possible. Development programmes for larger fields typically are less likely to be driven by purely financial considerations and are more likely to be effected by a policy of maximising ultimate recovery rates.
in 2007, this increase represents an additional annual loss of capacity through natural decline of around 700 kb/d. The implication is that an additional 700 kb/d of gross capacity — the equivalent of almost one-and-a-half projects the size of Khursaniyah in Saudi Arabia — had to be brought on stream in 2007 simply to offset the higher rate of natural decline compared with five years before. Case studies of how individual fields have behaved in the absence of any major investment provide a way of verifying the extent to which natural decline rates deviate from observed rates. In reality, there are few cases where a field has been left to decline over a long period without any capital spending whatsoever. Some cases can, nonetheless, be identified. The collapse of the Soviet Union caused upstream investment virtually to dry up for several years in the first half of the 1990s and led to a precipitous drop in oil production. During 1990-1995, the production-weighted average annual decline rate for the 19 largest Russian fields (with reserves in excess of 1 billion barrels) that were in decline at that time was close to 14%, though decline rates were certainly exaggerated by a lack of spending on operation and maintenance, too. Among those fields, Samotlor — the world’s sixth-largest in terms of initial reserves — experienced a year-on-year production decline of more than 16% over the same period. Among all the North Sea fields in our dataset, we have identified only nine that were not subject to any major development programme over a period of at least three years (the majority of fields have been developed in a continuous fashion, with sustained capital spending over much of the life of the field). The (arithmetic) average decline rate is 13.7% for all the fields in this sample (Table 10.13), which is very close to the natural decline rate we estimate for all North Sea fields (which make up virtually all of Europe’s production) based on 2002-2007 data. Decline rates among the nine fields range from under 6% to over 20%.
Field
Average year-on-year decline rates for selected North Sea oilfields Country
Initial oil reserves (million barrels)
Period of production without significant investment
Average decline rate*
Fulmar
UK
583
1997-2006
13.2%
Miller
UK
345
1998-2000
23.1%
Murchison
UK
332
1985-1989
20.3%
Ninian
UK
1 310
1994-1996
13.2%
Tern
UK
287
2003-2007
13.9% 12.0%
Thistle
UK
433
1996-2002
Auk
UK
197
1998-2003
5.7%
Denmark
298
2004-2007
11.9%
Netherlands
275
1969-1976
Skjold Rijswijk Average (arithmetic)
10.2% 13.7%
* Calculated as the cumulative average annual rate of decline between the first and last year of production over the specified period without significant investment. Source: Official government data; IEA analysis.
246
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
Table 10.13
The apparent inverse relationship between field size and the natural decline rate mirrors that between the rate of depletion of recoverable reserves (measured by the ratio of remaining reserves to production, R/P) and the decline rate. The R/P ratio is inversely correlated with the natural decline rate at the regional level, even though we have only eight regional data points. The four regions with the lowest R/P ratio — OECD North America, Europe and the Pacific, together with Asia — have the highest natural decline rates, while the Middle East, with the highest R/P ratio, has the lowest decline rate (Figure 10.11). In short, the natural decline rate appears to rise over time as a producing region matures and the R/P ratio drops.
Natural decline rate
Figure 10.11
Natural decline rates and reserves-to-production ratios by region, 2007
20% Pacific
18% 16%
North America
14%
Europe
12%
Asia
10%
Latin America Africa
8%
R2=0.6452
E. Europe/Eurasia
6% 4%
10
Middle East
2% 0% 0
10
20
30
40
50 60 70 80 Remaining reserves/production ratio (years)
Note: Natural decline rates are production-weighted. Sources: IHS databases; IEA databases and analysis.
Long-term prospects for natural decline rates
There is little reason to suppose that, for a given type and size of field in a specific location, the natural decline rate will change significantly in the future. However, a change in the mix of fields that will be developed in the future, including a shift towards smaller reservoirs and offshore deepwater fields, would be expected to drive up natural decline rates over time in all regions. On the other hand, a larger share of new field developments over the projection period is expected to come from onshore locations in the Middle East, where natural decline rates are the lowest (mainly because the average size of fields is high). This factor will offset, at least partially, the effect of declining field size on the weighted-average natural decline rate worldwide. Chapter 10 - Field-by-field analysis of oil production
247
© OECD/IEA, 2008
The scale of upstream investment required to match oil production to demand in the medium to long term hinges critically on the evolution of natural decline rates. The rate and type of investment, both in existing oilfields and in new fields, that will come on stream over the projection period will determine the extent to which observed decline rates diverge from natural rates.
We have assessed, by region, how average natural decline rates weighted by production could change in the future, using our Reference Scenario projections of crude oil and NGLs production, and additions to proven and probable reserves through reserves growth and discoveries (described in detail in Chapter 11). The correlation between the R/P ratio and the natural decline rate is applied to the projected R/P ratio in each region to derive an estimate of how the natural rate in each case might evolve between 2007 and 2030 (Figure 10.12). The results suggest that natural decline rates will tend to rise in all regions. At the world level, the increase in the production-weighted average decline rate over the projection period is about 1.5 percentage points, taking the rate to around 10.5% per year in 2030. The increase is particularly pronounced in North America, where the natural decline rate increases from about 14% to 17%, while the R/P ratio falls to about 10 years (as remaining reserves fall even faster than production). Projected change in natural decline rates and reserves-toproduction ratios by region, 2007 to 2030
Natural decline rate
Figure 10.12
22% 20% 18% 16% 14% 12% 10% 8% 6% 4% 2% 0%
2007
Pacific
2030 North America Europe World
Latin America
Asia Africa
0
10
20
E. Europe/Eurasia
30
40
Middle East
50 60 70 80 Remaining reserves/production ratio (years)
Note: Natural decline rates are production-weighted.
248
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
Sources: IHS databases; IEA databases and analysis.
CHAPTER 11
PROSPECTS FOR OIL PRODUCTION Running faster to stand still? H
I
G
H
L
I
G
H
T
S
In the Reference Scenario, world oil production (not including processing gains) rises by 26%, from 82.3 mb/d in 2007 to 103.8 mb/d in 2030. The share of OPEC countries in global output increases from 44% in 2007 to 51% in 2030. Saudi Arabia remains the world’s largest producer throughout the projection period, its output climbing from 10.2 mb/d to 15.6 mb/d. Worldwide, conventional crude oil production increases only modestly between 2007 and 2030 — by 5 mb/d — as almost all the additional capacity from new oilfields is offset by the decline in output at existing fields. Output from known oilfields that are already being developed or are awaiting development expands through to 2020, but then begins to drop, as few such fields are left to be brought into production and many of them enter their decline phase. Fields that are yet to be found account for about a quarter of total crude oil production by 2030. The bulk of the net increase in total oil production comes from natural gas liquids (NGLs), driven by the relatively rapid expansion in gas supply, and from nonconventional resources and technologies. Enhanced oil recovery, predominately from CO2 injection, makes a growing contribution. The regional breakdown of world oil production does not change dramatically over the projection period, but there are some important country-level changes. Saudi Arabia sees the largest incremental increase in oil production, followed by Canada. Brazil and Kazakhstan see large increases in output, while production in Russia falls. NGLs in Saudi Arabia and crude oil in Iraq dominate the increase in OPEC output.
OPEC crude oil and NGL production increases from 35.9 mb/d in 2007 to 44.0 mb/d in 2015 and to 52.0 mb/d in 2030, on the assumption that there are no major disruptions in supply and that the requisite investment occurs. The increase in crude oil and NGL output comes mainly from onshore fields: offshore production falls after 2015, despite rising output in Angola and Nigeria. The average size of onshore fields in Middle East OPEC countries declines over the Outlook period, especially after 2015.
Chapter 11 - Prospects for oil production
249
© OECD/IEA, 2008
Non-OPEC production of crude oil and NGLs declines from 44.8 mb/d in 2007 to 43.5 mb/d in 2015 and then falls further to 42.9 mb/d in 2030 in the Reference Scenario. Non-OPEC non-conventional oil production, in contrast, rises from 1.5 mb/d in 2007 to 7.9 mb/d in 2030.
These projections call for huge investments to explore for and develop more reserves, mainly to combat decline at existing fields. An additional 64 mb/d of gross capacity — the equivalent of six times that of Saudi Arabia today — needs to be brought on stream between 2007 and 2030. A faster rate of decline than projected here would sharply increase upstream investment needs and oil prices.
Global oil-production trends Summary of projections in the Reference Scenario In the Reference Scenario, world oil production rises (not including processing gains) by 26%, from 82.3 mb/d in 2007 to 103.8 mb/d in 2030. Declines in crude oil production at existing fields (those already in production in 2007) are more than compensated for by output from fields under or awaiting development and, mainly in the last decade of the Outlook period, fields that are yet to be found (Figure 11.1). Worldwide, production of conventional crude oil alone increases only modestly, from 70.2 mb/d to 75.2 mb/d over the period. The share of natural gas liquids (NGLs) and enhanced oil recovery (EOR), predominately from CO2 injection, in total oil production rises considerably, from 13% in 2007 to 25% in 2030. The contribution of non-conventional oil also rises substantially, from 2% in 2007 to 8.5% in 2030, with a marked spurt in the decade commencing now. Cumulative conventional oil production (crude and NGLs), which stood at 1.1 trillion barrels in 2007, is projected to rise to over 1.8 trillion barrels by 2030. Its share of currently estimated ultimately recoverable resources would, therefore, rise from slightly less than a third today to around one-half by 2030, though any increase in resources, for example, brought about by advances in technology, would result in a smaller increase in that share.
World oil production by source in the Reference Scenario
120
Natural gas liquids Non-conventional oil
100
Crude oil additional EOR
80
Crude oil - fields yet to be found
60
Crude oil - fields yet to be developed
40
Crude oil - currently producing fields
20 0 1990
250
2000
2010
2020
2030
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
mb/d
Figure 11.1
The projections take account of build-up and decline rates at existing fields and those that will come on stream over the projection period, new capacities from current, sanctioned and planned upstream projects, and potential discoveries over the projection period based on our country-by-country assessment of ultimately recoverable reserves (Box 11.1). Assumptions about decline rates are based on the analysis of Chapter 10. The share of world oil production from member countries of the Organization of the Petroleum Exporting Countries (OPEC), particularly in the Middle East, grows significantly, from 44% in 2007 to 51% in 2030. Their collective output of crude oil and NGLs grows from 35.9 mb/d in 2007 to 44.0 mb/d in 2015 and 52.0 mb/d in 2030 (Table 11.1). The contribution of NGLs to total conventional output in OPEC countries (and to a lesser extent elsewhere), rises quickly over the projection period, reaching 25% of OPEC output in 2030, compared with 13% in 2007. Saudi Arabia remains the world’s largest oil producer throughout the projection period, its production climbing from 10.2 mb/d in 2007 to 15.6 mb/d in 2030. Table 11.1
World oil production and supply in the Reference Scenario (million barrels per day) 2000
2007
2015
2030
66.0
70.2
73.0
75.2
29.0
31.1
35.9
38.9
6.7
9.2
10.8
7.4
37.0
39.1
37.1
36.3
Of which offshore NGLs OPEC Non-OPEC Non-conventional oil** OPEC Non-OPEC
15.9 7.8 3.0 4.9 1.2 0.2 1.1
15.2 10.5 4.7 5.7 1.6 0.1 1.5
15.4 14.4 8.1 6.4 4.6 0.4 4.2
16.3 19.8 13.2 6.6 8.8 0.9 7.9
Total production
Crude oil* OPEC Of which offshore Non-OPEC
75.0
82.3
92.0
103.8
OPEC
32.1
35.9
44.4
52.9
Non-OPEC
42.9
46.3
47.6
50.9
Processing gains
1.7
2.1
2.3
2.6
World oil supply
76.8
84.3
94.4
106.4
43%
44%
48%
51%
OPEC market share
11
Non-OPEC production increases much more slowly, from its current level of 46.3 mb/d to 47.6 mb/d in 2015 and 50.9 mb/d in 2030. Crude oil production alone is projected to decline slowly, from 39.1 mb/d in 2007 to 36.3 mb/d in 2030, as output from Chapter 11 - Prospects for oil production
251
© OECD/IEA, 2008
* Including condensates. ** Extra-heavy oil (excluding Venezuela), oil sands, chemical additives, gas-to-liquids and coal-to-liquids. Biofuels are not included.
offshore fields is unable to fully compensate for a steady drop in onshore production. The share of non-OPEC countries in world oil production declines by 7 percentage points, from 56% now to 49% in 2030. The reduction in non-OPEC production would be even more pronounced were it not for the growing contribution of oil from enhanced oil recovery (EOR) projects, which rises from 0.4 mb/d in 2007 to 3.6 mb/d in 2030.
Box 11.1
Modelling oil production in WEO-2008
For WEO-2008, trends in oil production to 2030 are modelled using a bottom-up methodology, making extensive use of the IHS database of worldwide proven and probable reserves and discovered fields, and country-by-country estimates of ultimately recoverable resources from the US Geological Survey (USGS). The methodology aims to replicate investment decisions in the oil industry by analysing the profitability of developing reserves at the project level (Figure 11.2). In the model, production in countries that are assumed to be open to private companies is separately derived, according to the type of asset in which investments are made: existing fields, new fields, non-conventional projects and exploration. Standard production profiles are applied to derive the production trend for existing fields and for those new fields (by country and type of field) which are brought into production over the projection period (see Chapter 10). Cash available for investments is a function of past production, the international oil price (taking account of the government take in oil revenues) and the investment policies of oil companies. Investment is made only in profitable projects. New projects generate cash, which is then spent on future profitable projects.
Investment in exploration, which is assumed to absorb a fixed share of total cashflow, determines the number of exploration wells drilled (based on assumptions about the cost of drilling). For each country, the amount of oil discovered (onshore and offshore) each year is based on an historical analysis which allows us to determine the relationship between the cumulative number of exploration wells and the total amount of oil found (known as a “creaming curve”).
252
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
The profitability of each type of project is based on assumptions about the capital and operating costs of different types of projects, the hurdle price (expressed as a proportion of the actual oil price assumed in the Reference Scenario) and the discount rate, representing the cost of capital. The net present value of the cash-flows of each type of project is derived from a standard production profile, which is determined by the size of the field and its location onshore or offshore. Projects are prioritised by their net present value and available cash is spent on the most potentially profitable projects, with total capital spending constrained by the total available cash-flow of the industry.
It was further assumed that total discoveries equal the ultimately recoverable resources, as estimated by the USGS. The main assumptions used in the model are the international oil price, exploration and development costs, the marginal tax take of the host country, and the rate of re-investment of cash-flows by oil companies. The price used to evaluate the profitability of projects is assumed to be a function of the assumed international market price. We assumed the current fiscal regimes and the average rates of re-investment observed by the oil industry in 2007 to remain constant over the projection period. We also assume in the Reference Scenario that upstream costs level off in 2009 and remain flat thereafter in real terms. The oil-supply projections for countries which are considered to be closed to international investments are based on exogenous assumptions about investment, on a country-by country basis, derived from announced plans and policies, as well as our judgment of the feasibility of the investment being made. Here, too, we apply standard production profiles to the fields which are assumed to be developed. The countries in question include Saudi Arabia, Kuwait, Iran, Iraq, Qatar, United Arab Emirates, Mexico and Venezuela — all of which, bar Mexico, are members of OPEC.
Chapter 11 - Prospects for oil production
253
11
© OECD/IEA, 2008
The projected growth in global oil production is significantly lower than in last year’s Outlook (but so, too, is demand). Total oil production in 2030, including non-conventional supplies (but not biofuels) is almost 10 mb/d lower than in WEO-2007. This reflects the slower economic growth assumption and the impact on demand of the large upward adjustment in our oil-price assumption. This oil price is, in turn, based on a re-assessment of the prospects for upstream investment in the main resource-rich countries, in light of the large escalation of project costs (see Chapter 13). It now looks much less likely that the key producing countries, in particular, will be willing and able to expand capacity as much and as quickly as previously assumed. Thus, this year’s Reference Scenario resembles closely the Deferred Investment Scenario of WEO-2005 (IEA, 2005). As a result, a bigger share of production will need to come from non-OPEC countries, where decline rates and development costs are generally much higher (see Chapters 10 and 13). Combined with sharp increases in upstream capital and operating costs, higher prices will be needed to bring forth the investment needed to expand capacity and to balance supply with demand. As always, future investment and decline rates are uncertain. The results of an analysis of the sensitivity of global oil markets to changes in these factors are presented in the last section of this chapter.
254
Reserves
Onshore Shelf Deepwater Non-conv.
Identified yet-to-be developed fields
Figure 11.2
© OECD/IEA, 2008
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
New discoveries
Existing fields (from start of projection)
Price threshold for economic analysis Existing fields net output
New fields production
Creaming curves based on USGS estimates of ultimately recoverable resources
Decline rate (aggregate)
Standard production profiles
Selected projects’ gross capacities
Technical and drilling costs
Potential greenfield projects
Available cash constraint
Architecture of the Oil Supply Model
Realised market price
Total production
Number of exploration wells
Industry cash-flows Industry total investments
Industry investment policy (re-investment rate)
Industry’s share of cash-flows (after government take)
Total cash
Exploration investments
Producing fields investments
Developments investments
Data input Assumption Variable Model output
Crude oil output at existing fields Our oil-production projections are based on a rigorous analysis of trends in output at existing fields (defined as those that were producing in 2007), which is also applied to known fields yet to be developed and fields yet to be found. Among existing fields, some are still building up, some are at plateau and the rest are in decline (see Chapter 10). Thus, the average year-on-year fall in the aggregate production of these fields will tend to accelerate over time, as more and more of them enter their decline phase. In the Reference Scenario, we assume that the decline rate (year-onyear) for all oilfields, once they have passed their peak, is constant for given types and sizes of fields in each region. However, the expected shift in the sources of crude oil, in terms of region, location and field size, means that the average productionweighted observed post-peak decline rate tends to rise, from 6.7% at the start of the projection period to 8.6% by the end.1 In total, output of crude oil from existing oilfields (not including non-conventional projects) drops from 70 mb/d in 2007 to 51 mb/d by 2015 and 27 mb/d by 2030 — a fall of 43 mb/d (excluding projected production using EOR). As a result, 64 mb/d of new capacity, in gross terms, needs to be brought on stream between 2007 and 2030 in order to maintain capacity at the 2007 level and meet the projected 21 mb/d increase in demand. The gross new capacity required by 2015 is about 30 mb/d. Oil production from existing oil sands and extra-heavy oil projects typically declines much less rapidly, as output is largely determined by the accessibility of the deposits and the intensity of mining or steam injection (see Chapter 9). Production from existing offshore fields declines much more rapidly than that from onshore fields, for the reasons outlined in Chapter 10. On average, output from existing onshore fields falls at a year-on-year rate of 3.2%, from 46 mb/d in 2007 to 22 mb/d in 2030. Output drops much more quickly at existing offshore fields, by an average of 6.3% per year, from 24 mb/d to 5 mb/d. The overall average annual fall in output at existing fields is proportionately much smaller in OPEC countries, at 3.3%, than in non-OPEC countries, where it is 4.7%, reflecting the fact that most OPEC fields are onshore. Production at existing fields falls by 17 mb/d in OPEC countries over 2007-2030, compared with 26 mb/d in non-OPEC countries (Figure 11.3).
11
1. See Chapter 10 for detailed estimates of decline rates.
Chapter 11 - Prospects for oil production
255
© OECD/IEA, 2008
The biggest fall in crude oil output from existing fields, in absolute terms, occurs in the Middle East, where some 11 mb/d of capacity needs to be replaced (Figure 11.4). This simply reflects the fact that this region is, by far, the biggest single producing region: the rate of the overall fall in production — an average of around 3% per year — is actually lower than that in any other region. In OECD Europe and OECD Pacific, offshore fields account for most of the loss of oil output; in Asia and Africa, the losses are evenly spread between offshore and onshore fields. Russian and other eastern European/Eurasian countries account for 17% of the overall 43 mb/d fall in output.
mb/d
Figure 11.3
Crude oil production from existing fields in OPEC and non-OPEC countries in the Reference Scenario
80
Non-OPEC - offshore
70
Non-OPEC - onshore
60
OPEC - offshore
50
OPEC - onshore
40 30 20 10 0 2007
mb/d
Figure 11.4
2015
2030
Crude oil production decline of existing fields by region in the Reference Scenario, 2007-2030
12
Offshore
10
Onshore
8 6 4 2 0 Middle East
E. Europe/ OECD North Eurasia America
Africa
Asia
Latin America
OECD Europe
OECD Pacific
Contribution of new fields to crude oil production
New conventional oil production will come both from known fields (included in proven and probable reserves) that are awaiting development and from fields that have yet to be found (as a result of exploration) but that are expected to be developed before 2030. The contribution of production from fields now awaiting development is projected to be 256
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
Despite continuing investment in the world’s existing fields, their aggregate output begins to decline already in 2008, as declines at mature fields outweigh increases at fields recently brought on stream. Output from new conventional oilfields, not yet in production, makes the biggest contribution to compensating for this loss of capacity and meeting rising global oil demand, with the remainder coming from non-conventional sources and from enhanced oil recovery (at both existing fields and those that will be developed during the projection period).
the larger of the two, even in 2030, but their share of gross capacity additions falls quickly towards the end of the projection period. Overall output from those fields peaks soon after 2020 as more and more of them are brought on stream, mature and enter their decline phase. By contrast, the contribution from fields yet to be found grows quickly after 2015, as more of them are found and are selected for development, in some cases ahead of yetto-be-developed known fields (Figure 11.5). By 2030, output from yet-to-be-found fields (19 mb/d) approaches that from known but yet-to-be-developed fields (23 mb/d).
mb/d
Figure 11.5
World crude oil production from new fields in the Reference Scenario
45
Yet to be found
40
Yet to be developed
35 30 25 20 15 10 5 0 2010
2015
2020
2025
2030
11
Known but yet-to-be-developed fields
Production from these yet-to-be-developed fields is projected to rise steadily to a peak of 29 mb/d just after 2020 and then fall back to around 23 mb/d in 2030 (Figure 11.6). This means that 220 billion barrels out of the 257 currently identified, or 86%, are produced over the projection period. By 2030, the contribution from onshore fields is twice as large as that from offshore fields, although offshore production dominates until about 2020. In non-OPEC countries, known yet-to-be developed fields contribute 11 mb/d in 2030, after peaking at 16 mb/d. Offshore production represents over two-thirds of total output from these fields until about 2020. In OPEC countries, yet-to-be-developed fields produce 11 mb/d in 2030, slightly down on the peak of 14 mb/d.
2. IEA analysis based on IHS data.
Chapter 11 - Prospects for oil production
257
© OECD/IEA, 2008
At the end of 2007, world conventional oil reserves in fields that have been discovered but not yet developed amounted to an estimated 257 billion barrels.2 OPEC countries held 133 billion barrels, or just over half. About 65% of OPEC’s undeveloped conventional reserves are located onshore. In non-OPEC countries, onshore reservoirs account for only 38% of total reserves. Over half of the world’s undeveloped reserves are in the Middle East and Russia. The Middle East alone holds 99 billion barrels, nearly 40% of the world total.
mb/d
Figure 11.6
Crude oil production from yet-to-be-developed fields in OPEC and non-OPEC countries by location in the Reference Scenario
30
Non-OPEC - offshore
25
Non-OPEC - onshore OPEC - offshore
20
OPEC - onshore
15 10 5 0 2005
2010
2015
2020
2025
2030
Onshore yet-to-be-developed reserves are concentrated in the Middle East and Russia. Together, they account for three-quarters of all onshore fields awaiting development. Offshore reserves are more equally distributed worldwide (Figure 11.7). The majority of the fields awaiting development in Africa, North America and eastern Europe/Eurasia and nearly all of them in Latin America and OECD Europe and Pacific are offshore.
Billion barrels
Figure 11.7
Conventional proven and probable crude oil reserves in yet-to-be-developed fields by region, end-2007
100
Offshore Onshore
80 60 40 20 0 Russia
Africa
Latin America
OECD Other E. North Europe/ America Eurasia
Asia
OECD Europe
OECD Pacific
The 257 billion barrels of yet-to-be-developed conventional oil reserves are distributed in 1 874 fields, 971 of which are onshore and 903 offshore (Figure 11.8). Two-thirds of these fields are in non-OPEC countries. Africa holds 21% of them, roughly evenly distributed between offshore and onshore. All but six of the 141 fields awaiting development in OECD Europe are offshore. Some 93% of Russia’s yet-to-bedeveloped fields are onshore.
258
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
Middle East
Number of fields
Figure 11.8
Number of yet-to-be-developed oilfields by region and location, end-2007
400
Offshore
350
Onshore
300 250 200 150 100 50 0 Middle East
Russia
Africa
Latin America
OECD Other E. North Europe/ America Eurasia
Asia
OECD Europe
OECD Pacific
The average size of yet-to-be-developed fields tends to be much larger in the Middle East, at 335 million barrels, compared with the three OECD regions, Asia and Africa (Table 11.2). The average size of offshore fields in the Caspian region is high, because of the Kashagan field in Kazakhstan, which is due to come on stream in 2014. The average field size in OPEC countries is twice that of fields in non-OPEC countries. Average size of yet-to-be-developed oilfields by region, end-2007 (million barrels)
11
Onshore
Offshore
Overall average
OECD North America
28
102
62
OECD Europe
64
52
53
OECD Pacific
33
48
47
Russia
160
570
187
Other E. Europe/Eurasia Asia Middle East Africa Latin America OPEC Non-OPEC *Excluding Kashagan. Sources: IEA analysis based on IHS data.
116 58 324 74 36 213 84
854 (228*) 60 368 104 232 182 119
310 (142*) 59 335 91 152 201 103
Yet-to-be-found fields Conventional oil production from yet-to-be-found fields is projected to reach 19 mb/d in 2030, based on the projected discovery of 114 billion barrels of reserves worldwide over the projection period. Almost 11 mb/d comes from offshore fields (Figure 11.9). Onshore production comes mostly from OPEC countries — 8 mb/d out of a total of Chapter 11 - Prospects for oil production
259
© OECD/IEA, 2008
Table 11.2
8.7 mb/d in 2030 — as most undiscovered resources in the Middle East are onshore (USGS, 2000). By contrast, the bulk of production from yet-to-be-found offshore fields comes from non-OPEC countries: they produce 7.9 mb/d out of a total of 10.7 mb/d in 2030. Non-OPEC offshore yet-to-be-found fields are about equally divided between Russia and other Eurasian countries, Africa and OECD North America.
Enhanced oil recovery (EOR) Production using EOR technology is projected to contribute an additional 6.4 mb/d 3 to world oil supply in 2030, with most of the increase occurring after 2015 (Figure 11.10). Three-quarters of this increase comes from just four countries: the United States, Saudi Arabia, Kuwait and China (in ranked order). Cumulative production in 2007-2030 amounts to 24 billion barrels, out of an estimated potential of about 300 billion barrels — of which one-third is in Saudi Arabia (see Chapter 9). Most of the new EOR projects that are expected to be implemented during the projection period involve CO2 injection. In total, about 9.8 gigatonnes of CO2 is captured and stored in CO2-EOR oil projects over the projection period.4 These projects are exclusively onshore, as the technology for offshore CO2-EOR is not expected to be sufficiently advanced for it to be deployed before the end of the projection period. Current EOR technology typically recovers about an additional 10% of oil in place. A bigger and more widely applied carbon penalty than assumed in the Reference Scenario (see Chapter 1) would significantly increase the potential for EOR.
mb/d
Figure 11.9
World crude oil production of yet-to-be-found oilfields in OPEC and non-OPEC countries by location in the Reference Scenario
20
Non-OPEC - offshore Non-OPEC - onshore
16
OPEC - offshore OPEC - onshore
12 8 4
2010
2015
2020
2025
2030
3. Over and above the estimated 2.5 mb/d of EOR production from existing fields — see Chapter 9 (Table 9.2). 4. This projection assumes that about 2.5 barrels of oil are recovered per tonne of CO2 injected. 260
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
0 2007
Figure 11.10
Enhanced oil recovery by country in the Reference Scenario
Oman
2015
Algeria
2030
Qatar United Arab Emirates Russia Canada Kuwait China Saudi Arabia United States 0
0.5
1.0
1.5
2.0
2.5 mb/d
Natural gas liquids (NGLs) Output of natural gas liquids — light hydrocarbons that exist in liquid form underground and that are produced together with natural gas and recovered in separation facilities or processing plants — is expected to grow rapidly over the Outlook period. Global NGL production is projected to almost double, from 10.5 mb/d in 2007 to just under 20 mb/d in 2030. This increase is driven by the steady rise in natural gas output (see Chapter 12). The bulk of the increase comes from OPEC countries, where gas production (to supply local markets and new LNG projects) is projected to expand quickest. OPEC NGL production almost triples, from 4.7 mb/d in 2007 to over 13 mb/d in 2030. The Middle East accounts for four-fifths of this increase. Non-OPEC NGL production increases by about 1 mb/d, to close to 7 mb/d in 2030 (Figure 11.11). These projections assume that the average NGL content of gas production is constant over the projection period.
mb/d
Figure 11.11
11
World natural gas liquids production by OPEC and non-OPEC countries in the Reference Scenario
20
Non-OPEC OPEC
15
10
0 1990
2000
Chapter 11 - Prospects for oil production
2007
2015
2030
261
© OECD/IEA, 2008
5
Non-conventional oil Non-conventional oil — extra-heavy oil (excluding Venezuela), oil sands, chemical additives, gas-to-liquids and coal-to-liquids — is expected to play an important role in offsetting the decline in production from existing fields and supplementing supplies from new conventional oilfields and NGLs. The global supply of non-conventional oil is projected to increase from 1.7 mb/d in 2007 to 8.8 mb/d in 2030 (Figure 11.12). Canadian oil-sands projects make, by far, the largest contribution, increasing by 4.7 mb/d, followed by coal-to-liquids (CTL) projects in the United States and China, which together increase by 1 mb/d. Output from gas-to-liquids (GTL) projects expands from about 50 kb/d in 2007 to 650 kb/d in 2030. These projections need particular qualification, especially those for gas- and coal-to-liquids, given the uncertainties about future choices of technology and the resulting mix of fuels, the evolution of the technologies, how production costs will develop relative to alternative ways of exploiting the resources and future environmental constraints.
mb/d
Figure 11.12
World non-conventional oil production by type in the Reference Scenario
9
Additives*
8
Oil sands – mining
7
Oil sands – in situ
6
Extra-heavy oil
5
Gas to liquids
4
Coal to liquids
3 2 1 0 2007
2010
2015
2020
2030
*Methyl tertiary butyl ether (MTBE) and other chemicals.
Production of oil from Canadian oil sands is projected to increase from 1.2 mb/d in 2007 to 5.9 mb/d in 2030. Total proven recoverable oil-sands reserves are estimated at 175 billion barrels, of which 35 billion barrels can be mined and 140 billion barrels can be produced in situ (see Chapter 9). All the mineable reserves are projected to be developed over the projection period, with production from mining projects reaching about 1.4 mb/d in 2030. Current mining techniques are not economic if the depth of the oil sands exceeds about 75 metres; since most oil-sands reserves lie at depths which require the use of in situ steam-assisted production techniques, almost all of the projected increase in oil-sands production comes from in situ projects. Of the more than 50 billion barrels that are projected to be developed to 2030, only about 30 billion barrels derive from projects that are already sanctioned, planned or at the feasibility study stage. Our projections of oil-sands production are inevitably uncertain in view of the environmental challenges, notably water needs and CO2 emissions (see Chapter 9). 262
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
Oil sands and extra-heavy oil
Extra-heavy oil production projects are expected to be concentrated in Venezuela (such as the Carabobo project), as this country holds the major part of the relevant world resources. However, due to the IEA practice of classifying extra-heavy oil in Venezuela as conventional oil, Venezuelan extra-heavy oil production is not included here. Production of extra-heavy oil outside Venezuela is projected to occur mainly in Kuwait and in a few isolated projects in Brazil, Vietnam and Italy, and may reach more than 0.7 mb/d in the last decade of the projection period. Gas-to-liquids (GTL) GTL technology involves converting natural gas into liquid fuels with longer-chain hydrocarbons, such as diesel fuels that can be directly used to fuel diesel-engine vehicles, naphtha and chemical feedstock. GTL technology generally uses the principles of the Fischer-Tropsch (F-T) process, in which methane-rich natural gas is transformed, via steam-reforming or partial oxidation, into a mixture of carbon monoxide and hydrogen, called “syngas”. Catalytic conversion converts the syngas into diesel fuels. While the F-T process was developed in the 1920s, significant progress has been made on the design of catalysts. The trend is to move from iron to cobalt and other metals. Three plants — Sasol’s Mossel plant in South Africa, Shell’s Bintulu plant in Malaysia and the Oryx plant in Qatar (a joint venture between Qatar Petroleum and Sasol) — account for the bulk of the 50 kb/d of GTL capacity currently in operation worldwide. Two plants are under construction: Shell’s 140-kb/d Pearl project in Qatar, the first train of which is due on stream by the end of the current decade and the second by 2012, and Chevron/ Nigerian National Petroleum Corporation’s 34-kb/d Escravos project in Nigeria, expected on stream by the beginning of the next decade. They will bring total world capacity to above 200 kb/d by around 2012. All these plants will produce primarily diesel, together with smaller volumes of naphtha.
11
In the Reference Scenario, fewer GTL projects are brought on stream than in last year’s Outlook. Planned plants in Egypt and Australia are assumed to proceed, but others, including that proposed in Qatar but delayed because of a moratorium on new gas projects, do not proceed. World GTL production reaches 650 kb/d in 2030. Chapter 11 - Prospects for oil production
263
© OECD/IEA, 2008
The long-term prospects for GTL hinge critically on costs. Costs have dropped sharply in the past, from $120 000 per barrel per day of capacity in the mid-1950s to less than $40 000 in 2000 — partly due to economies of scale. But costs have rebounded in recent years, with the surge in capital and operating costs that has afflicted all sectors of the oil and gas industry and the higher cost of the gas feedstock. At present, each barrel of diesel produced requires 8 000 to 10 000 cubic feet of gas. Current GTL production costs per barrel are now in the range of $40 to $90, depending on the price of the feedstock gas. GTL plants generate between 0.2 and 0.25 tonnes of CO2 per barrel produced. If a CO2 penalty of $50 per tonne were introduced, production costs would rise by $10.00 to $12.50 per barrel of synfuel. How the industry might develop in the longer term is uncertain. Recent sharp increases in construction costs, technical difficulties in commissioning the Oryx plant and a tightening of LNG supply in the near to medium term may lead to a shift in strategy towards LNG in many gas-producing countries. GTL is a highly capital-intensive chemical process, involving more complex engineering and operation, compared with LNG.
Coal-to-liquids (CTL) CTL — a well-established technology that transforms coal into oil products — is expected to play a small but growing role in meeting the world’s fuel needs. The CTL process requires access to coal of stable quality, but has been given a new lease of life by technical improvements and by higher oil and gas prices. Global production of CTL is projected to grow from a mere 0.13 mb/d today to 1.1 mb/d in 2030. In addition to one plant that has been operating in South Africa for many years, a number of projects are assumed to be developed in China and the United States, because of their abundance of low-cost resources. India is not assumed to begin production of CTL before 2030, though plans to do so have been discussed for several years. The uncertainties surrounding CTL prospects are very large, because of technical, economic and environmental considerations. There are two main types of CTL technology in use or under development (IEA, 2006): The two-stage liquefaction process, which uses gasification of coal to produce syngas and then a F-T process to generate the fuel. It has been commercially used in South Africa since 1955. The Secunda plants there currently produce the equivalent of 130 kb/d of fuels and chemicals, using three different types of catalyst (lowtemperature iron, high-temperature iron and low-temperature cobalt). The plants generally produce 20% to 30% of naphtha, along with 70% to 80% of diesel, kerosene and fuel oil. The process uses 0.35 to 0.45 tonnes of coal per barrel of oil produced. New plants have an overall energy efficiency of around 40%. The more energy-efficient direct liquefaction process involves the dissolution at high temperature and pressure of a high percentage of coal in a solvent, followed by catalysed hydrocracking. The chemistry of this process is more complex. Output includes 20% to 30% of naphtha and 70% to 80% of diesel and liquefied petroleum gas. The first commercial project using the technology is the 20-kb/d Shenhua project located in Inner Mongolia (China), now nearing completion, with a planned capacity expansion to 100 kb/d. The process uses 0.3 to 0.4 tonnes of coal per barrel of oil produced.
A big drawback with CTL processes is their very high CO2 emissions: each barrel of oil produced gives rise to 0.5 to 0.7 tonnes of CO2. A 50-kb/d plant, accordingly, generates an average 11 Mt of CO2 per year. On this basis, a penalty of $50 per tonne of CO2 would add $20 to $30 per barrel to production costs. Large CTL projects will need access not only to large coal reserves (typically at least 2 billion tonnes), but also to CO2 storage sites if CCS is deployed. Plans to produce up to 30 Mt per year (around 600 kb/d) of fuels from CTL in China by 2020 have been announced. But environmental concerns, including emissions and 264
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
Average CTL capital costs for a 50-kb/d unit are estimated at around $85 000 per barrel per day of installed capacity (Rahim, 2008), with the gasification unit representing the largest cost component (30% of the total cost). Scale matters as both unit operating and capital costs fall by more than 30% with an increase in capacity from 10 to 80 kb/d. Cost factors vary regionally: costs in China are about 30% lower than the United States. Current costs for a plant in China with 1 million tonnes (Mt) per annum production capacity are $40 to $60 per barrel of oil produced.
access to water, along with spiralling costs and coal prices, have led the Chinese government to impose stricter rules for plant construction and operation. In the United States, several CTL projects have been announced, totalling more than 300 kb/d, but more are still at the feasibility study stage.
Crude oil quality There has been a small shift in the average quality of crude oil produced over the past decade towards slightly heavier and higher-sulphur grades (Eni, 2006). The shift towards heavier crudes is expected to continue in the longer term, but not as fast as projected in previous WEOs. Indeed, average API gravity5 is expected to rise through to 2011, mainly due to increased supplies of condensate and a preponderance of lighter (and sweeter) crudes coming from Saudi Arabia’s capacity-expansion programme (IEA, 2008). The fall in the average API gravity of world crude oil production resumes in 2012 and accelerates after 2015. The share of light and ultra-light grades (with an API gravity of at least 35°) in total crude oil production is projected to fall from 26.1% in 2005 to 24.4% in 2015 and 22.7% in 2030. The share of medium grades (with an API gravity of between 26° and 35°) increases by one percentage point from 53.7% in 2005 to 54.6% in 2030. Heavy grades (with an API gravity of 10° to 26°) account for 16.1% of world production in 2030, up from 12.8% in 2005. The average sulphur content of crude oil worldwide is expected to rise appreciably in the longer term, from the current levels of close to 1.14%. Figure 11.13
Crude oil quality worldwide in the Reference Scenario
100%
11
Unassigned Heavy
80%
Medium Light and ultra-light
60% 40% 20% 0% 2005
2015
2030
The near doubling of production of NGLs over the projection period helps to offset, in part, the projected modest fall in the average gravity of crude oil. NGLs include a mixture of propane, butane, pentane, ethane and other light hydrocarbons in widely varying proportions. The average API degree of these liquids is typically at least 60°. 5. The American Petroleum Institute gravity, or API gravity, is a measure of the relative density of petroleum liquids compared with water, expressed in degrees. The higher the degree, the lighter the oil.
Chapter 11 - Prospects for oil production
265
© OECD/IEA, 2008
Source: IEA analysis based on Eni (2006).
Outlook by country and region There is no dramatic shift in the regional breakdown of world oil production over the projection period in the Reference Scenario, but there are some important changes at the country level (Figure 11.14). Saudi Arabia sees the largest incremental increase in oil production, followed by Canada. Eurasian countries, notably Kazakhstan, see a large increase in crude oil output, while production in Russia declines. Saudi Arabia and Iraq dominate the increase in OPEC (and Middle East) production, with most of the additional capacity coming from NGLs in Saudi Arabia and from crude oil in Iraq. In the United States, an increase in the production of non-conventional oil largely offsets a decline in conventional crude oil production. Figure 11.14
Change in oil production by country/region, 2007-2030 Crude oil
Other non-OPEC Norway United Kingdom Other Middle East Mexico Russia India United States China Other OPEC UAE Venezuela Nigeria Other OPEC Middle East Kuwait Iran Brazil Other E. Europe/Eurasia Iraq Canada Saudi Arabia
Natural gas liquids Non-conventional
–4
–2
0
2
4
6 mb/d
Non-OPEC production
Investment in exploration will be crucial to increasing production in non-OPEC countries. The majority of the existing offshore fields in non-OPEC countries will be in the final decline phase of their production profile by 2030, resulting in a 12-mb/d drop in output compared with today. In the medium term, the shortfall is met from known but yet-to-be-developed fields, but output even from those fields is expected 266
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
Non-OPEC production of crude oil and NGLs in the Reference Scenario declines from 44.8 mb/d in 2007 to 43.5 mb/d in 2015 and then falls further to 42.9 mb/d in 2030 (Table 11.3). Non-conventional oil production, in contrast, rises from 1.5 mb/d in 2007 to 7.9 mb/d in 2030. Most of the decline in conventional production is a result of falling output levels in OECD countries, but that same group accounts for three-quarters of the increase in non-conventional oil output.
to start to decline in aggregate by around 2020, from a peak of about 16 mb/d to about 11 mb/d in 2030. Output from offshore fields that are yet to be found starts to fill the gap from 2015 and reaches 8 mb/d in 2030. Table 11.3
Non-OPEC oil production in the Reference Scenario (million barrels per day) 2000
2007
2015
2030
Crude oil and NGLs
41.8
44.8
43.5
42.9
OECD
20.9
18.0
15.0
14.1
North America
13.3
12.5
11.1
11.4
2.1
2.1
1.9
1.9
3.5 7.7 6.8 3.3 2.7 0.9 20.9
3.5 6.9 4.9 2.6 1.7 0.6 26.8
2.4 6.8 3.3 2.0 0.9 0.6 28.5
3.0 6.5 2.1 1.3 0.5 0.6 28.9
E. Europe/Eurasia
8.1
12.9
14.3
16.5
Kazakhstan
0.7
1.4
2.4
4.3
Russia
6.5
10.1
10.4
9.5
Asia
5.6
6.4
5.8
5.1
China
3.2
3.7
3.3
3.5
India
0.7
0.8
0.8
0.5
Middle East
2.1
1.7
1.4
1.1
Africa
1.9
2.3
2.0
1.7
Latin America
3.2
3.5
5.0
4.5
Canada Mexico United States Europe Norway United Kingdom Pacific Non-OECD
Brazil Non-conventional* OECD
1.3
1.8
3.5
3.4
1.1
1.5
4.2
7.9
0.9
1.3
3.6
6.7
Canada
0.6
1.2
3.3
5.9
United States
0.3
0.1
0.3
0.6
Non-OECD Total
0.2
0.2
0.5
1.2
42.9
46.3
47.6
50.9
Of which NGLs
4.9
5.7
6.4
6.6
OECD
3.7
3.7
3.9
3.9
Non-OECD
1.2
2.0
2.5
2.8
11
Onshore production in non-OPEC countries, excluding EOR, is also projected to fall, from 24 mb/d in 2007 to 16 mb/d in 2030 (Figure 11.15). Known but yet-to-bedeveloped fields contribute about 6 mb/d and yet-to-be-found fields 1 mb/d by the end of the projection period. NGLs rise modestly from 5.7 mb/d in 2007 to a little under 7 mb/d in 2030, reflecting flat gas production in non-OPEC countries. Chapter 11 - Prospects for oil production
267
© OECD/IEA, 2008
* Extra-heavy oil, oil sands, chemical additives, gas-to-liquids and coal-to-liquids. Biofuels are not included.
mb/d
Figure 11.15
Non-OPEC oil production by type in the Reference Scenario
60
Additives* Extra-heavy oil
50
Gas to liquids 40
Coal to liquids Oil sands
30
Natural gas liquids 20
Crude oil additional EOR
10
Crude oil - offshore Crude oil - onshore
0 2007
2015
2030
* Methyl tertiary butyl ether (MTBE) and other chemicals.
In the longer term, most of the additional offshore production will come from about 150 known fields, each holding between 50 and 600 million barrels of proven and probable reserves. In total, they hold about 20 billion barrels, in both shallow offshore and deep water. Again, Brazil has the largest offshore potential, with more than 50 fields containing a total of some 8 billion barrels — equal to 40% of total non-OPEC known offshore undeveloped reserves. Continuing active exploration, notably in the pre-salt layer of the Santos Basin, is likely to boost Brazil’s proven reserves substantially in the coming years. The United States has about 15 undeveloped fields, holding a total of 1.5 billion barrels. Kazakhstan holds about 1 billion barrels in 5 fields, each of more than 50 million barrels. Mexico, Malaysia, Norway and the United Kingdom together hold similar potential of about 1 billion barrels in about 5 to 7 fields. They also possess about 30 other discovered fields with reserves of less than 50 million barrels. The potential for offshore oil production in Russia is limited both by the number of 268
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
Offshore fields account for most of the gross additions to conventional oil capacity in non-OPEC countries over the Outlook period. About 180 sanctioned, planned and proposed projects, to develop an estimated 65 billion barrels of proven and probable reserves, are projected to yield new production capacity over the next five to six years. Most of the offshore projects are small: only 44 of them concern fields holding more than 300 million barrels. The bulk of non-OPEC offshore production in this period, including both shallow offshore and deep water, comes from a few major projects, including Tupi in Brazil, the fifth phase of Sakhalin in Russia, Kashagan in Kazakhstan, Jidong Nanpu in China and ACG in Azerbaijan. Brazil and the United States (Gulf of Mexico) are the only non-OPEC countries with more than two major projects in the pipeline. Other smaller-scale offshore developments include Ku Maloob Zaap in Mexico, Kikeh and Gumusut-Kakap in Malaysia and Moho-Bilondo in the Republic of Congo. The overall paucity of projects in the near term is mostly due to the current high costs of developing offshore fields, low levels of investment in exploration over the past decade and unfavourable regulatory and fiscal frameworks.
known fields (less than 10) and by volume (less than 1 billion barrels in total). Several discoveries have been made in Australia, Canada, Republic of Congo, Gabon, India, Thailand and Vietnam; but the average size of the fields is well below 100 million barrels, so their potential contribution to non-OPEC offshore production is limited at the prices assumed in the Reference Scenario and at current costs. Developing groups of small fields simultaneously could lower costs and make more of them economically viable. In contrast to the relative abundance of viable offshore projects, prospects for onshore projects in non-OPEC countries, outside of Russia, are very poor. Beyond 2015, the majority of known but yet-to-be-developed onshore fields with reserves of more than 100 million barrels are located in Russia, where the physiographic location of the fields complicates their exploitation considerably. Mexico and China hold most of the rest of the onshore potential in non-OPEC countries in fields holding more than 100 million barrels. As for smaller fields holding between 50 and 100 million barrels, fewer than 100 await development. Exploration will undoubtedly yield many new discoveries, but the maturity of the resource base in most non-OPEC regions suggests that the potential for finding fields that are large enough to be exploited profitably is fairly limited. In the Reference Scenario, yet-to-be-found onshore fields in non-OPEC countries contribute less than 1 mb/d in 2030, compared with 8 mb/d from offshore fields. Oil production in the United States is expected to pick up slightly in the next few years, mainly thanks to a surge in output in the Gulf of Mexico. About 5 billion barrels of proven and probable reserves in the Gulf are expected to be developed before 2015. A major project, Thunder Horse, came on stream in the third quarter of 2008 — about three years later than originally planned. It is projected to produce over 200 kb/d at peak by the beginning of the next decade. Other fields, such as Atlantis, Kaskida, Tahiti, Shenzi, Great White, Puma and Cascade, add another 400 kb/d by 2015. The average size of these fields is about 200 million barrels — about 50% of the average field size in Brazil. They are located in deep water and, therefore, are likely to have high post-peak decline rates (see Chapter 10). Onshore, the Nikaitchuq field with about 200 million barrels of reserves, is also expected to be developed. The additions to US output from these projects compensate only temporarily for declines at existing fields in Alaska and in the lower 48 states. Production in the longer term is expected to resume its long-term downward trend, reaching 6.5 mb/d in 2030. The lifting of the moratorium on offshore drilling proposed by the current administration — not assumed in the Reference Scenario because Congress had not yet approved it at the time of writing — would temper the decline, but only marginally.
11
Mexico’s proven oil reserves have fallen continuously over the past two decades, mainly because of under-investment in exploration. In 2007, Mexico produced around 3.5 mb/d of oil. Production at Cantarell, Mexico’s largest field and the third-largest Chapter 11 - Prospects for oil production
269
© OECD/IEA, 2008
Canadian production will become increasingly dominated by oil sands in Alberta. Conventional crude oil production is projected to continue to decline, but this is largely offset by increased output of NGLs; the combined output edges down from 2.1 mb/d in 2007 to 1.9 mb/d in 2030. Oil-sands output, in contrast, jumps from 1.2 mb/d to 5.9 mb/d.
offshore field in the world by initial reserves, is falling sharply. Year-on-year declines averaged only 5% per year between the field’s peak in 2003 and 2007, but output has reportedly dropped precipitously since 2007: at end-2007, it was 16% down on the beginning of the year. According to Pemex, the national company and sole operator of all Mexican fields, output at Cantarell is set to fall by a further 16% by end-2008. Total Mexican crude oil production is projected to bottom out at around 2.4 mb/d by 2015, as the Ku Maloob Zap and Chicontepec fields build up and reach their plateau. It then recovers gradually to more than 3 mb/d in the last decade of the projection period. Mexico holds 3 billion barrels of reserves in onshore fields which are yet to be developed and the above figures are boosted by output from new offshore discoveries — on the assumption that exploration is stepped up. In OECD Europe, production from mature North Sea basins will continue its long-term decline. Output is projected to drop to 3.3 mb/d by 2015 and 2.1 mb/d in 2030. Despite higher oil prices, European exploration and development have declined sharply in the last few years, as the prospects for discovering significant new reservoirs have deteriorated. Of more than 28 projects planned in the North Sea, only a handful hold more than 100 million barrels. The larger ones include the Lochnagar and Callanish/ Brodgar fields in the United Kingdom, which are projected to produce a bit less than 100 kb/d until 2015, and the Goliat, Tyrihans, Alvheim and Skarv/Idun fields in Norway, with aggregate production of slightly less than 200 kb/d. Most of the other fields being developed are extensions to earlier developments.
In Russia, production is expected to level off in the near term. The most significant onshore projects are Vankorskoye (with 3 billion barrels of proven and probable reserves), Priobskoye (1.6 billion barrels), Verkhne-Chonskoye (1.2 billion barrels) and Talakanskoye (900 million barrels). All of these fields are due to start producing oil before 2013, adding a total of 1.1 mb/d from around 2015. They start to decline by around 2020. Other important fields holding significant reserves in the region of 500 to 600 million barrels are also expected to be developed, including Tagulskoye and Suzunskoye (660 million barrels), Vladimir Filanovski (600 million barrels), YurubchenoTokhomskoye (560 million barrels), Salym (500 million barrels) and Khylchuyuskoye Yuzhnoye (480 million barrels). The additional contribution from those fields is close to 500 kb/d between 2015 and 2020. Beyond 2015, Russia has a substantial potential to develop new projects, with numerous fields reported to hold between 200 and 500 million barrels. Russia has more large onshore fields than any region bar the Middle East. Nonetheless, relatively rapid declines in mature fields in western Siberia and the Volga-Ural region are expected to offset the production increases from new developments. Long-term production prospects depend to a large extent on government licensing and fiscal policies, as well as on how quickly investment in pipeline and export infrastructure is forthcoming. Recent fiscal changes are expected 270
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
Prospects for finding more oil in Australia are good. Projected output growth there is supported in the short term by production from the Angel and Pyrenees fields, and in the medium term by the Vincent and Enfield fields. Total production levels off at about 0.5 mb/d in 2015, but grows to 0.6 mb/d in the second half of the projection period with new discoveries, although on average they are modest in size.
to stimulate more upstream investment, but are not sufficient to sustain production growth in the longer term. We project total Russian oil output to drop to 9.5 mb/d in 2030, after peaking at 10.4 mb/d around 2015. In other Eurasian countries, new export routes are expected to stimulate production growth. In Kazakhstan, there is potential for growth at the three main fields, Tengiz, Kashagan and Karachaganak. Capacity expansions at the Tengiz and Karachaganak fields, with combined reserves of more than 3 billion barrels, add about 500 kb/d of peak capacity. When the Kashagan field comes on line (at the earliest in 2013), combined production from these three fields reaches more than 1.5 mb/d, with Kazakhstan entering the small group of countries producing more than 2 mb/d. Production from new oilfield developments in China will not be sufficient to offset declining production from mature onshore areas, including the super-giant Daqing field, which has been in production since the 1960s. Production from the Changqing area, Tarim Basin and Xinjiang is expected to remain flat or grow modestly in the near term. Chinese production dips to 3.3 mb/d in 2015 but rebounds to 3.5 mb/d by 2030. Output is expected to be broadly flat in the rest of Asia through to the end of the current decade and to decline thereafter. Recent discoveries in India, however, including the Mangala (430 million barrels), Bhagyama (150 million barrels) and Aishwariya (105 million barrels) fields, delay the decline until 2015. These three fields produce between 100 and 150 kb/d from 2012 to 2020. Rehabilitation of the Bombay High field in India will also contribute. Some new developments are expected in Vietnam, Malaysia and Thailand, boosting their collective output marginally.
Production in Latin America is projected to grow from 3.5 mb/d in 2007 to 5 mb/d in 2015 and then to fall to 4.5 mb/d in 2030. Brazil, the region’s largest non-OPEC producer, has proven oil reserves of 8.5 billion barrels (not including some large recent finds that have yet to be officially confirmed) and huge potential for further discoveries. After a first pilot phase, which is projected to see production reaching about 100 kb/d before 2015, the giant Tupi field discovered in 2006 is assumed to reach plateau production of 1 mb/d before 2020, assuming that the technical challenges in producing oil from the sub-salt layer can be overcome (see Chapter 9). Ultimate capacity could be higher, depending on ultimate recoverable reserves (currently estimated tentatively at 5 to 8 billion barrels). Other fields, including the Jupiter and Carioca fields announced in 2008, could add to production eventually, though further appraisal work is needed before formal development plans can be prepared. In the shorter term, 13 fields, including the Papa-Terra, Marlim Leste, Marlim Sul, Jubarte, Cachalote and Jabuti, holding between 300 and 700 million barrels, are currently Chapter 11 - Prospects for oil production
271
11
© OECD/IEA, 2008
Prospects for non-OPEC African countries are good to 2010, but production declines thereafter, to 1.7 mb/d in 2030, down from 2.3 mb/d today. Côte d’Ivoire, Republic of Congo, Equatorial Guinea and Sudan are expected to achieve higher output through to 2010 (see Chapter 15). The most significant project in these countries is the MohoBilondo in Republic of Congo, which is projected to produce about 100 kb/d at plateau in the next three years. Further production growth will depend on the success of exploration and appraisal drilling in firming up reserves, and on how favourable the investment and legislative frameworks are.
being developed. Each of these fields is due to start producing between 50 kb/d and 200 kb/d before 2015. They add about 1.3 mb/d to Brazilian production around 2015. Brazil alone holds more than 20% of the total non-OPEC offshore reserves that are expected to be developed over the medium term.
OPEC production OPEC crude oil and NGL supply increases from 36 mb/d in 2007 to 44 mb/d in 2015 and to 52 mb/d in 2030 (Table 11.4). These projections assume that oil prices remain high by historical standards. By 2030, OPEC countries account for 51% of world oil production. Crude oil production increases from 31.1 mb/d in 2007 to 38.9 mb/d in 2030. EOR projects contribute 2.8 mb/d to this increase: 1.2 mb/d in Saudi Arabia, 800 kb/d in Kuwait, 320 kb/d in the United Arab Emirates, 240 kb/d in Qatar and 210 kb/d in Algeria. EOR is projected to provide one-third of the net increase in OPEC crude oil output. Offshore OPEC production declines, from around 9 mb/d in 2007 to 7.4 mb/d in 2030, after peaking at just below 11 mb/d around 2015. Onshore production (including EOR) is projected to increase from 22 mb/d in 2007 to 32 mb/d in 2030. OPEC oil production in the Reference Scenario (million barrels per day) 2000
2007
2015
2030
Crude oil and NGLs
32.0
35.9
44.0
52.0
Middle East
21.3
23.6
30.3
37.1
Iran
3.8
4.4
4.5
5.4
Iraq
2.6
2.1
3.0
6.4
2.2 0.9 9.3 2.6 10.7 1.4 0.7
2.6 1.3 10.2 3.1 12.2 2.2 1.7
2.9 2.0 14.4 3.7 13.6 2.0 2.3
3.3 2.5 15.6 3.9 15.0 2.3 2.6
Kuwait Qatar Saudi Arabia UAE Non-Middle East Algeria Angola Ecuador
0.4
0.5
0.3
0.2
Indonesia
1.4
1.0
0.8
0.3
Libya
1.5
1.9
2.0
2.2
Nigeria
2.2
2.3
3.4
3.7
Venezuela
3.1
2.6
2.7
3.6
Non-conventional*
0.2
0.1
0.4
0.9
32.1
35.9
44.4
52.9
Of which NGLs
3.0
4.7
8.1
13.2
Middle East
1.8
3.0
6.0
9.8
Non-Middle East
1.2
1.7
2.0
3.4
Total
* Extra-heavy oil (not including Venezuela), oil sands, chemical additives, gas-to-liquids and coal-to-liquids. Biofuels are not included.
272
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
Table 11.4
In OPEC countries, a decline occurs in output from existing fields of about 17 mb/d, from 31 mb/d in 2007 to less than 15 mb/d in 2030 (excluding the impact of EOR). Output from existing onshore fields drops by one-half and from offshore fields by three-quarters. These production losses are made good mainly by output from known fields yet to be developed. Production in these fields, however, is expected to peak at about 14 mb/d in 2020 and to fall back to 11 mb/d by 2030. Yet-to-be-found fields produce about 1 mb/d in 2015 and 11 mb/d in 2030. Whether the investment needed to bring these fields into production will be made is one of the biggest uncertainties in this Outlook. Output of OPEC NGLs is projected to triple, from almost 5 mb/d in 2007 to over 13 mb/d in 2030, reflecting huge upstream gas developments to meet rapidly rising domestic demand and to supply new LNG projects. Non-conventional oil is expected to remain very limited. Excluding extra-heavy oil in Venezuela,6 total output is projected to grow from 0.1 mb/d to 0.9 mb/d in 2030. The main contribution to non-conventional OPEC oil production comes from Middle East countries, specifically extra-heavy oil in Kuwait (0.6 mb/d) and GTL in Qatar (about 0.2 mb/d).
mb/d
Figure 11.16
OPEC oil production by type in the Reference Scenario
60
Additives*
50
Extra-heavy oil Gas to liquids
40
Natural gas liquids
30
Crude oil additional EOR
20
Crude oil - offshore
11
Crude oil - onshore
10 0 2007
2015
2030
There are 101 upstream oil projects currently under development and planned in OPEC countries as a whole, involving an estimated 48 billion barrels of proven and probable reserves. Of these projects, 56 are onshore, involving about 28 billion barrels of reserves. The largest element of new production is due to come from Saudi Arabia, which will bring four new onshore fields into production (see below). In the longer term, there is considerable potential for other known onshore fields to be brought into production — in stark contrast to the situation in most non-OPEC countries. There are more than 100 onshore fields awaiting development, each holding more than 100 million barrels and with combined reserves of more than 50 billion barrels. The bulk of them are in just three countries: Saudi Arabia, Iran and Iraq. Two-thirds of those reserves are concentrated in about 30 giant fields (each holding more than 6. In this Outlook, extra-heavy Venezuelan crude oil is categorised as conventional oil.
Chapter 11 - Prospects for oil production
273
© OECD/IEA, 2008
* Methyl tertiary butyl ether (MTBE) and other chemicals.
500 million barrels), including Sharar, Niban, Jaladi, Dhib and Lugfah fields in Saudi Arabia, Halfayah, Hamrin, Tuba and Ratawi fields in Iraq, and Hosseinieh, Kuskh and Kuh-I-Mand fields in Iran. We project that 70 fields in these three countries will be brought into production before 2030. Their development costs per barrel of production are on average the lowest in the world. Over the next five to six years, 45 offshore projects are either under development or firmly planned in OPEC countries, totalling 20 billion barrels. The largest project by far is the Manifa field in Saudi Arabia, accounting for one-quarter of these reserves. This field is expected to come on line in 2012 and will produce close to 1 mb/d when the plateau is reached about three years later. Angola and Nigeria together account for two-thirds of the remaining 15 billion barrels. Nigeria ranks second in terms of total reserves developed over the medium term, with about 5 billion barrels in 11 projects, all of which involve more than 500 million barrels of reserves. Other large projects include Corocoro (450 million barrels) in Venezuela, Reshadat and phases 9 & 10 of South Pars in Iran (400 million barrels each), phase 2 of the Al Khaleej project in Qatar (380 million barrels) and Jeruk 2 (280 million barrels) in Indonesia. No new developments in the United Arab Emirates are expected in the medium term but expansions of existing fields such as Upper Zakum, Lower Zakum, Umm Shaif and Rumaitha are planned there. Several other offshore fields that are awaiting development could be brought on line before 2030. The offshore fields in OPEC Middle East countries generally are more economic to develop than those in other OPEC countries as they are mainly in shallow water and their average size is high (about three to four times higher than in Angola and Nigeria). Saudi Arabia will continue to play a vital role in balancing the global oil market. Its willingness to make timely investments in oil-production capacity will be a key determinant of future price trends. Five major onshore projects, Khurais, Khursaniyah, Hawiyah, Shaybah and Nuayyim, which collectively hold 13 billion barrels of reserves, are all in the final phases of development and are projected to provide a total gross capacity addition of close to 3 mb/d by 2015. Expansions of the Zuluf, Safaniyah and Berri fields could give further momentum to Saudi Arabian production.7 There are 6.2 billion barrels of reserves in 11 other known fields in Saudi Arabia that have yet to be developed — the highest in the world. Each of them holds reserves of at least 150 million barrels and the average is 450 million barrels. The largest known undeveloped field, Hasbah 1, holds 1.8 billion barrels. Part of the increase in gross capacity will be offset by the decline in capacity at existing fields. Saudi Arabia aims to maintain spare capacity in the range of 1.5 mb/d to 2 mb/d in the long term.
7. At the Jeddah Summit of 22 June 2008, Oil Minister Al-Naimi said Saudi Arabia could expand capacity to more than 15 mb/d, subject to future market conditions. This increment was centered on five key fields: Zuluf, Safaniyah, Berri, Khurais and Shaybah.
274
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
The prospects for oil production in Iraq remain very uncertain. Output is beginning to recover following the war in 2003. Iraqi production has always been low relative to the size of the country’s reserves. Iraq holds the world’s third-largest remaining oil reserves, after Saudi Arabia and Iran, with 115 billion barrels, according to OPEC. Iraq undoubtedly has the potential to increase its production capacity by several
million barrels per day. On the assumption that security across the country continues to improve, we project production to rise steadily from 2.1 mb/d in 2007 to around 3 mb/d in 2015 and to accelerate to 6.4 mb/d in 2030. This projection implies a five- to six-fold increase in investment in the upstream oil sector. Achieving this will hinge on political stability and providing opportunities for foreign oil companies on satisfactory commercial and fiscal terms. Oil and NGL production potential in Iran is also very large. Output falls to 4.1 mb/d in 2010, largely because of delays in implementing new projects, but recovers to 4.4 mb/d by 2015 and jumps to 5.4 mb/d in 2030. A total of 15 onshore projects is due to come on stream over the next five to six years. Leaving aside the Pars 12 and Darkhovin III fields, which have reserves of more than than 600 million barrels, the 13 other projects have reserves between 150 and 300 million barrels. Five additional projects are reported to hold less than 100 million barrels. The sum of the reserves being developed in these 20 projects ranks second among OPEC countries, with exploitation of more than 4.5 billion barrels due to start in this period. Total production from the projects, which are at or close to the final phase of development, is projected to reach about 1 mb/d on average in 2010-2020. It is nonetheless very uncertain that these production levels will be achieved, given limitations on investment and geopolitical factors. Developments in Kuwait are concentrated on the Sabriyah and the Umm Niqa fields, where a total of 50 kb/d of natural gas liquids capacity is being added, with completion expected in 2011. Another project at the Sabriyah field adds 160 kb/d of crude oil production by 2015. The reserves being developed are estimated to total 2.5 billion barrels. A water treatment project at the Greater Burgan field — the second largest in the world — is expected to boost gross capacity by 120 kb/d by 2009. Beyond 2020, we project Kuwait to develop about 6 billion barrels of heavy and extra-heavy oil resources. In addition to the 3.3 mb/d from conventional sources in 2030, the extra 0.5 mb/d from those projects are counted as non-conventional in our Reference Scenario. Total Kuwaiti oil production is projected to reach 3.8 mb/d in 2030.
11
Nigeria’s proven oil reserves are estimated at 25 billion barrels. The country produced 2.3 mb/d of oil in 2007. In addition to the Agbami field, which started producing this year, several fields are currently under development, the most important of which are the Bonga Southwest/Aparo, Akpo, Bonga Northwest and Usan and Egina — all of them in deep water. They are projected to add 1.2 mb/d of gross capacity by 2015, boosting total production to around 3.6 mb/d (taking account of declines at existing fields). Investment risk in Nigeria is high, because of political instability and corruption. Chapter 11 - Prospects for oil production
275
© OECD/IEA, 2008
Outside the Middle East, Venezuela has the largest reserves among OPEC countries and the greatest long-term potential for expanding production. Several large onshore projects underway involve extra-heavy oil, including the Carabobo project, which we project to start producing by around 2015. This project produces close to 1 mb/d when it reaches full capacity between 2018 and 2020. Other projects, in the Junin, Petro Cedeno and Machete areas, target reserves of around 1 billion barrels each. Over the longer term, there is strong potential for new extra-heavy oil projects like Carabobo, if investment can be mobilised. The prospects for developments of conventional oilfields are much more limited. The La Ceiba onshore and Corocoro offshore fields together are projected to produce around 100 kb/d by around 2020.
Nigeria is seeking an increase in its share of the OPEC production ceiling. Considering its huge potential and recent government reforms to oil and gas legislation, we expect the country to produce some 3.7 mb/d by 2030. Angola is leading the momentum in developing offshore production. Some 15 projects are expected to come on line before 2015. All but one is located in deep water. The combined reserves of these projects are 5.5 billion barrels. Except in the case of the Pazflor, which alone holds some 1 billion barrels, other projects usually include several neighbouring fields, such as Gindungo/Canela/Gengibre and Plutao/Saturno/ Venus/Marte. Total production from this group of projects is projected to reach 1 mb/d around 2012-2013. It then inches up to a bit less than 1.5 Mb/d in 2015 and dips below 1 mb/d beyond 2020. Offshore potential from Angola lies in 26 fields, with more than 50 million barrels each and total reserves of 3 billion barrels. Unless some of those fields are geographically close enough to be developed simultaneously, their average size of about 110 million barrels makes their development uncertain from an economic point of view, as most of them are located in deep waters. Libya is considered a highly attractive oil province because production costs are low, it is close to Europe and it has well-developed infrastructure. Libya produces highquality, low-sulphur crude oil. Two major projects are underway: the re-development of the Sirte basin and an expansion of the Nafoora field. Ten smaller projects over the next four to five years have an average reserve level of 120 million barrels. Oil output in Libya is projected to increase from 1.9 mb/d now to 2.2 mb/d by 2030. Algeria and Indonesia have the fewest and, on average, smallest current projects among the OPEC countries. In both countries, the biggest project involves the development of about 350 million barrels of reserves. There are just two projects underway in Indonesia and three in Algeria, with combined reserves of less than 1 billion barrels.
Sensitivity of oil output to decline rates
In order to demonstrate the importance to oil-production prospects of the decline rates of existing fields, and to quantify the extent to which changes in decline rates affect investment needs and oil prices, we have developed two variants of the Reference Scenario: a high decline-rate case, in which the decline rate for all fields producing in 2007 is assumed to be one percentage point higher than in the Reference Scenario throughout the projection period, and a low decline-rate case, in which the rate is 276
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
The production decline rate that will prevail over the projection period will determine the extent of the need to make new investments in existing fields and new fields: the faster the decline-rate, the bigger the investments that will need to be made, and vice versa. Our Reference Scenario decline-rate assumptions are derived from a detailed bottom-up approach, examining the situation in each country separately on a fieldby-field basis, and drawing on extensive consultation with the oil industry, as well as assumptions about production and investment policies. Nonetheless, the level of present decline rates and of future trends is extremely uncertain. Decline rates could deviate significantly from assumed trends, with major implications for oil markets.
assumed to be one percentage point lower. All other exogenous variables are assumed to be unchanged, including the decline-rates and production profiles of new fields. Crucially, we have assumed that the level of global oil supply and demand is unchanged over the Outlook period, in order to be able to quantify the change in price that would correspond to a higher or lower decline rate, and the consequent change in investment needs: for example, in the high decline-rate case, higher prices would be needed to make more projects profitable and bring forth more investment. The differences in the international oil prices and investment in the two cases, vis-à-vis the Reference Scenario, are large. In the high decline-rate case, the oil price in 2030 is almost $20 per barrel higher than in the Reference Scenario, reaching more than $140 (Table 11.5). Cumulative investment in the upstream oil industry over the whole projection period, at $6.1 trillion, is over $1 trillion, or one-fifth, higher than in the Reference Scenario. The gap is already significant in the period to 2015, as more exploration and development spending is needed to compensate for the faster drop in production from existing fields. Table 11.5
World cumulative investment in the upstream oil sector under different decline-rate assumptions for existing fields, 2007-2030 Oil price* in 2030 ($2007/barrel)
2007-2015
2016-2030
122 141 100
1 819 2 179 1 817
3 217 3 923 2 531
Reference Scenario High decline-rate case Low decline-rate case
Investment (billion $2007) 2007-2030 5 036 6 102 4 349
11
* Average IEA crude oil import price.
Chapter 11 - Prospects for oil production
277
© OECD/IEA, 2008
In the low decline-rate case, there is a very small difference in 2015, compared with the Reference Scenario, as most of the investment over that period has already been sanctioned. Over the projection period, however, as the gap widens between production from existing fields and demand, the marginal cost of producing each barrel of new supplies rises, so that investment needs increase faster than the additional capacity needs. As discussed in detail earlier in this chapter, this reflects the change in the composition of the supply base in the longer term, as more offshore fields and more fields smaller in size than those today need to be developed.
© OECD/IEA, 2008
CHAPTER 12
NATURAL GAS RESOURCES AND PRODUCTION PROSPECTS Who will supply the world’s growing gas demand? I
G
H
L
I
G
H
T
S
Global proven reserves of natural gas at the end of 2007 stood at close to 180 trillion cubic metres — equal to around 60 years of current production. Like oil, gas resources are highly concentrated in a small number of countries and fields. Three countries — Russia, Iran and Qatar — hold 56% of the world’s reserves, while just 25 fields hold almost half. OPEC countries also hold about half. Remaining reserves have more than doubled since 1980, with the biggest increases coming from the Middle East. Reserves have risen by more than 15% even since 2000, despite rising annual production. As with oil, the bulk of the increase in reserves in recent years has come from upward revisions for fields that have been producing or have been undergoing appraisal and development work. Although the size of gas discoveries has been steadily declining in recent decades (in the same way as for oil), discoveries continue to exceed production. The USGS estimates remaining ultimately recoverable resources of conventional natural gas, including proven reserves, reserves growth and undiscovered resources, at 436 tcm. Cumulative production to 2007 amounts to 13% of total initial resources. Non-conventional gas resources — including coalbed methane, tight gas sands and gas shales — are much larger, amounting perhaps to over 900 tcm, with 25% in the United States and Canada combined. Worldwide, gas resources are more than sufficient to meet projected demand to 2030. The outlook for production by region depends largely on the proximity of reserves to markets, which primarily determines the cost of supply. Much of the world’s gas resources is located far from the main markets, so that only a small proportion of the economic potential has yet been exploited. In the Reference Scenario, world natural gas production is projected to rise from just over 3 tcm in 2007 to 4.4 tcm in 2030. Production expands in all major regions, with the exception of OECD Europe, where production is already in decline, and in OECD North America, where output begins to dip in the second half of the Outlook period. Output grows most strongly, in both volume and percentage terms, in the Middle East, tripling between 2007 and 2030 to 1 tcm per year. Most of this increase comes from Iran and Qatar. Production in Latin America and Africa also grows briskly. It is very uncertain whether all of the investment needed to achieve the projected production levels in these regions, especially the Middle East and Russia, will materialise.
Chapter 12 - Natural gas resources and production prospects
279
© OECD/IEA, 2008
H
Gas resources and reserves Conventional gas Global resources of natural gas that can be economically extracted are of a similar magnitude (on an energy-equivalent basis) to those of oil. Proven reserves1 at the end of 2007 stood at 179 trillion cubic metres according to Cedigaz, an international centre for gas information, and 175 tcm according to the Oil and Gas Journal (O&GJ) — equal to around 60 years of current production. Other sources show similar estimates.2 OPEC countries hold about half and OECD countries well under 10%. Remaining reserves have more than doubled since 1980, with the biggest increases coming from the Middle East (Figure 12.1). Reserves have risen by about 12% since 2000, despite rising annual production.
Proven reserves of natural gas
Trillion cubic metres
Figure 12.1 200
Latin America North America
160
Asia/Pacific Africa
120
Europe and Eurasia 80
Middle East
40 0 1980
1990
2000
2007
Source: BP (2008).
1. For definitions, which are the same as those for oil, see Chapter 9. 2. BP (2008) shows a figure of 177 tcm.
280
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
Natural gas reserves, like those of oil, are unevenly dispersed across the world, being highly concentrated in a small number of countries and fields. Three countries — Russia, Iran and Qatar — hold 56% of the world’s remaining proven reserves, while just 25 fields contain 48% of the global total. The Middle East as a whole holds 41% of reserves and Russia alone one-quarter. Regional shares in global production are very different from shares in reserves: the Middle East, for example, accounts for only 11% of world production. North America holds only 4.5% of the world’s reserves, but 26% of production (Figure 12.2). These disparities largely reflect differences in the proximity of reserves to markets and the investment climate.
Figure 12.2
Regional share in natural gas production and proven reserves, 2007
50%
Share of production Share of proven reserves
40% 30% 20% 10% 0% Latin North Asia Africa E. Europe/ Middle America America Eurasia East Note: 2007 data are provisional for the OECD and are estimates for the non-OECD countries. Pacific
Europe
Sources: Reserves - Cedigaz (2008); production - IEA databases.
As with oil, the bulk of the increase in proven reserves in recent years has come from upwards revisions for fields that have been producing or have been undergoing appraisal and development work. Nonetheless, newly discovered volumes remain large. Although the size of gas discoveries has been steadily declining in recent decades, unlike oil, volumes discovered continue to exceed production (Figure 12.3). Natural gas discoveries* and cumulative production, 1960-2006
10
250
8
200 150
6 4
100
2
50
12
Discoveries Production Cumulative discoveries (right axis) Cumulative production (right axis)
0
0 1960-1969 1970-1979 1980-1989 1990-1999 * Additions to proven reserves from new fields.
Trillion cubic metres
Trillion cubic metres per year
Figure 12.3
2000-2006
Estimates of ultimately recoverable conventional resources, including cumulative production, remaining proven and probable reserves, reserves growth and undiscovered resources, vary widely. In its 2000 assessment, the US Geological Survey estimated Chapter 12 - Natural gas resources and production prospects
281
© OECD/IEA, 2008
Sources: IHS and IEA databases.
ultimately recoverable resources at at 436 tcm in its mean case (USGS, 2000). Adjusted for cumulative production and changes in reserves since that assessment, remaining resources (including proven reserves, undiscovered resources and reserves growth) at end-2007 are estimated at 380 tcm (2 240 billion barrels of oil equivalent), of which about half have already been proven (Table 12.1). Cumulative production to 2007 amounts to 13% of total initial conventional gas resources (compared with 33% for conventional oil). These numbers do not take account of the latest USGS estimates of the resource potential of the Arctic region, released in July 2008 (USGS, 2008). According to that assessment of 25 Arctic provinces, the mean undiscovered gas resources are about 46 tcm (excluding NGLs), and more than 70% of the overall undiscovered gas potential is in three provinces (West Siberian Basin - 18 tcm; East Barents Basins - 9 tcm; and Arctic Alaska - 6 tcm). Table 12.1 Region
Remaining proven reserves and mean estimates of ultimately recoverable conventional resources of natural gas, end-2007 Remaining proven reserves (tcm)
Proven reserves share (%)
Ultimately recoverable resources (URR) (tcm)
Reserves as share of URR (%)
OECD North America
8.0
4.5
62.5
12.8
Latin America
7.7
4.3
24.5
31.5
Europe
6.2
3.5
25.5
24.3
53.8
30.1
132.2
40.8
Former Soviet Union Africa
14.6
8.2
33.9
43.1
Middle East
73.2
41.0
134.8
54.2
Asia-Pacific World
15.2
8.5
29.9
50.8
178.7
100.0
443.3
40.3
Sources: Reserves — Cedigaz (2008); resources — USGS (2000); IEA estimates.
3. Extremely tight gas reservoirs may have recovery factors lower than 10%. 4. The 61% recovery rate was calculated from a sample of 8 560 fields in 2006. An earlier study by the same author, using 1997 data, yielded a recovery rate of 75%.
282
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
Recovery rates — the share of the estimated amount of hydrocarbons initially in place that is economically recoverable — are significantly higher for conventional gas fields than for oil fields. They are thought to range from 30% to close to 100%,3 averaging around 61%, based on IHS data (Laherrere, 2006).4 Average rates are likely to increase with technological improvements, such as the increased use of well-stimulation techniques that target stranded and bypassed pockets of gas. Use of enhanced gas recovery (EGR) with the injection of CO2 from anthropogenic sources is another option, but this technology is in its infancy compared with enhanced oil recovery (EOR). The world’s first pilot project of CO2-EGR is currently being operated by GDF Suez in the offshore K12B field in the Netherlands.
Box 12.1
Sour gas reserves: a costly nuisance?
Sour gas reservoirs contain, in addition to methane, significant quantities of hydrogen sulphide (H2S) and/or carbon dioxide (CO2). More than 40% of the world’s gas reserves are estimated to be sour (Table 12.2). In the Middle East, 60% of proven gas reserves are sour. Southeast Asia’s largest gas reservoir is the Indonesian Natuna field, with 1.3 tcm of gas reserves, but it has a CO2 content as high as 71%. Technical difficulties have delayed Natuna’s development for more than two decades. The presence of H2S or CO2 poses a number of challenges for drilling and production. The majority of the completion components need to use special alloys to prevent corrosion. A bigger problem is dealing with the gases produced and avoiding their release to the atmosphere. In the case of CO2, re-injection and underground storage is a potential solution that has been successfully implemented. The Norwegian Sleipner project and the Algerian In-Salah project each store over 1 million tonnes of the gas each year. In 2008, the Snohvit project started injecting 0.7 million tonnes of CO2 per year into storage, while the planned Australian Gorgon project plans similar procedures to avoid emissions of more than 3 million tonnes of CO2-equivalent per year by 2010. Carbon capture and storage (CCS) in gas production is a near-term opportunity to demonstrate the CCS technology, as the cost of capture of CO2 from natural gas streams is significantly lower than from the flue gases of power stations (IEA, 2008). The cost of separating (from gas production), compressing, transporting and re-injecting CO2 is site-specific (generally in the range of $5 to $20 per tonne of CO2).
12
World proven sour gas reserves, end- 2006 High H2S only
High CO2 only
High H2S and CO2
Total
(tcm)
(tcm)
(tcm)
(tcm)
% of total reserves
Mexico & Latin America
0.3
1.1
0.3
1.7
Europe
0.1
0.7
0.3
1.1
21 19
Former Soviet Union
0.8
10.1
7.3
18.2
34
Africa
0.0
0.5
0.5
1.0
8
Middle East
2.6
0.4
40.9
44.0
60
Asia-Pacific
0.3
4.4
2.3
7.1
46
World
4.2
17.2
51.6
73.1
43
Note: Excludes North America. High H2S is more than 100 parts per million; high CO2 is more than 2%. Source: Bourdarot (2007).
Chapter 12 - Natural gas resources and production prospects
283
© OECD/IEA, 2008
Table 12.2
Non-conventional gas Non-conventional gas embraces a set of gas resources that are generally contiguous in nature, sometimes referred to as “resource plays” in the industry, that require special drilling and stimulation techniques to release the gas from the formations in which they occur. Non-conventional gas includes coalbed methane, tight gas sands and gas shales. Such resources are widespread worldwide, but their development has generally been limited so far to North America. Determining the amount of gas in place in non-conventional reservoirs is a complex task, due to the heterogeneous structure of the reservoirs and of the production profiles, which can differ significantly from conventional wells. Total worldwide non-conventional gas resources are estimated at over 900 tcm, with 25% in the United States and Canada combined, and 15% each in China, India and the Former Soviet Union. Non-conventional gas accounts for a significant and rising share of US gas production. Eight of the ten largest onshore gas fields discovered in the United States in the 1990s were non-conventional reservoirs, including the 1-tcm Newark East Barnett Shale and the Powder River coalbed methane deposits. The US Energy Information Administration projects that the share of gas production from non-conventional resources in total domestic production will increase from 40% in 2004 to 50% in 2030, with this gas making up 28% of the US natural gas supply increase between 2004 and 2030 (DOE/EIA, 2007). Tight gas sands Tight gas sands are generally defined as natural gas reservoirs with a continuous accumulation system5 and low permeability. World tight gas sand recoverable reserves are estimated at 200 tcm, with the largest quantities in the Americas (38%), Asia-Pacific (25%), the Former Soviet Union (13%), the Middle East and North Africa combined, and sub-Saharan Africa (11% each). The United States (principally) and Canada have been the leaders in tight gas sands development. In the United States in 2006, tight gas sands accounted for 40% of all the gas wells drilled and nearly 30% of total gas production. Although the number of tight gas sands wells drilled per year and production have been rising rapidly over the last decade in the United States (Figure 12.4), productivity, measured by gas reserves per well, has fallen by half. As a result, drilling activity needs to continue to rise if production is to be sustained. That factor, compounded by increasing upstream costs (see Chapter 13), has led to a near-quadrupling of production cost per unit of gas in the last five years.
The majority of the world’s coal resources, estimated at 86 to 283 tcm (Creedy and Tilley, 2003), are found at depths where coal cannot be mined. Coalbed methane is the methane contained in coalbeds that, due to the deposit’s depth or the coal’s poor quality, cannot be mined. In operating coal mines, methane gas is generally considered to be a hazard, as well as creating an environmental problem if vented to 5. An accumulation that is pervasive through a large and continuous area.
284
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
Coalbed methane
Tight gas drilling in the United States
16
70
14
60
12
50
10
40
8 30
6
1996-2000 2001-2002 2003-2005
20
4
10
2 0
Million cubic metres per well
Number of wells per year (thousand)
Figure 12.4
Tight gas sands
Coalbed methane Wells
Gas shales
Tight gas sands
Coalbed methane
Gas shales
0
Well productivity (right axis)
the atmosphere. Coalbed methane can be produced with well-bore technologies that are similar to those used for conventional hydrocarbons. However, production can be difficult where the formations are tight, requiring techniques such as hydraulic fracturing to enhance well productivity. Water in the porous space needs to be removed before the gas can be extracted, which may also complicate the production process, increasing capital costs and giving rise to environmental concerns.
Coalbed-methane production is also expanding outside the United States. In Canada, large resources are located in the Mannville, Ardley and Horseshoe Canyon formations. Over 10 500 wells had been drilled by early 2007 and production is expected to exceed 0.3 tcf (8.5 bcm) in 2008, making up more than 5% of Canada’s total gas production. In Australia, commercial production started in Queensland in 1998 and in 2005 supplied more than 7% of the country’s total gas needs. It is projected to provide 35% to 50% of Eastern Australia’s gas needs by 2020 (Geoscience, 2008). International oil companies, including Shell, Petronas and BG, have acquired coalbed-methane assets in Australia, some with the intention of exporting the gas as LNG. India is expected to produce up to 0.1 tcf (2.8 bcm) by 2010, accounting for nearly 5% of total Indian gas production. The Indonesian Chapter 12 - Natural gas resources and production prospects
285
12
© OECD/IEA, 2008
Despite these difficulties, the contribution of coalbed methane to natural gas supply is growing rapidly in the United States. Commercial coalbed-methane production started there in the late 1980s, spurred by tax incentives, and grew with new technologies and expansion to new basins. Advances in seismic technology, horizontal well drilling, hydraulic fracturing and basin modelling have helped reduce uncertainties over the assessment of reserves and their commercial exploitation. US production has expanded from the original San Juan Basin to several new basins in Colorado, New Mexico and Wyoming, reaching 1.8 trillion cubic feet (tcf) (50 bcm) in 2006 and accounting for nearly 10% of US non-associated gas production.
government hopes to develop the country’s coalbed-methane resources, offering better terms for production-sharing agreements than for conventional natural gas projects. Coalbed methane has been slower to take off in China, even though the country has the world’s third-largest reserves and despite the large amounts of methane released through venting from coal mines (estimated at over 0.2 tcf, or 5.7 bcm, per year) and the associated safety and environmental risks. Increasing production is a priority in China’s 11th five-year national plan (2006-2010), with the stated objective of producing nearly 0.3 tcf — 8.5 bcm — by 2010. Russia has over 80 tcm of recoverable coalbed-methane reserves, including more than 10 tcm in the Kuzbass basin, located at a depth of less than 1 200 metres. A few experimental projects have been started by Promgaz, a subsidiary of Gazprom, but production remains negligible. Other countries interested in developing their coalbed-methane reserves include Ukraine and Romania.
Box 12.2
CO2-enhanced coalbed methane (ECBM) production: a dual benefit?
The injection of CO2 into deep unmineable coal seams — a technology known as CO2-enhanced coalbed methane (ECBM) — can produce a double benefit: it can enhance production of coalbed methane and store CO2. The first application has been under consideration, along with nitrogen injection, for more than a decade. CO2 is preferentially adsorbed6 in coal (compared to CH4), because fractures in the structure of the coal can absorb twice as much CO2 as the methane it initially contained. Recent studies have shown that some low-quality coals in the United States could store five to ten times as much CO2 as the initial methane (Mastalerz, 2004).
6. Adsorbtion refers to the formation of a thin film on a surface through condensation.
286
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
Estimates for CO2 storage using ECBM range from 150 to 220 gigatonnes of CO2 (Table 12.3). The potential is biggest in the United States and Australia. Between 0.1 and 0.2 tonnes of methane are expected to be recovered for each tonne of CO2 injected. Commercial-scale injection of CO2 for ECBM has been undertaken in the US San Juan Basin, recovering 77% to 95% of the methane in place. Other experimental ECBM projects have also been launched in Canada, China, Japan and Poland. A number of technical questions need to be addressed before this technology can become commercial, including the interaction of CO2 with the formation rock, which can reduce permeability and lower productivity.
Table 12.3
Enhanced coalbed-methane CO2 sequestration potential by country/region
Country
Potential (Gigatonnes)
United States
35-90
Australia
30
Indonesia
24
Former Soviet Union
20-25
China
12-16
Canada
10-15
India
4-8
South Africa & Zimbabwe
6-8
Western Europe
3-8
Central Europe Total
2-4 146-228
Sources: Gale (2004); Reeves (2003).
Shale gas: a neglected resource becomes valuable Significant volumes of natural gas are sometimes contained in layers of shale rock, often hundreds of metres thick, in regions with conventional oil and gas resources. Until recently, producing the gas from such formations, usually discovered when drilling for oil and gas at deeper levels, was not profitable because of the low permeability of the rock. Shale gas can also occur in areas without commercial oil and gas resources. As exploration in those areas has been minimal, those resources may be significantly larger than currently thought.
12
Global shale-gas reserves are estimated at 450 tcm, with 35% in the Americas, 35% in the Asia-Pacific region and 15% in the Middle East (Rogner, 1997). The United States is currently the only large-scale producer, with output reaching 20 bcm in 2005. This is expected to rise to 28 bcm by 2010, accounting for nearly 5% of total US gas production. An economic analysis is underway of the potential development of shale gas in British Columbia, Canada: early tests with a horizontal well have achieved production rates of 5 million cubic feet per day per well; there are plans to drill up to 20 additional wells by the end of 2008. Chapter 12 - Natural gas resources and production prospects
287
© OECD/IEA, 2008
Natural gas can be stored in shales through different storage mechanisms: free gas in the pores of fractures and adsorbed gas. Due to the very tight nature of the shale, production requires stimulation (hydraulic fracturing): in shallow reservoirs, fluids mixed with nitrogen are often used; in deeper formations, low-viscosity fracturing fluids (slickwater) have replaced more expensive treatments (Frantz and Jochen, 2006). New methods of completing the wells involve the combined use of horizontal wells and a series of staged hydraulic fractures. The recent take-up of those techniques has made possible the development of the Barnett Shales in the United States. Recovery factors are expected to increase from an estimated 12% up to 50% during this decade.
Gas hydrates: are we getting closer to commercial exploitation? Methane or gas hydrates are the most abundant source of hydrocarbon gas on earth. They are crystal-like solids that are formed when methane is mixed with water at low temperature and moderate pressure. They occur at relatively shallow depths, usually offshore or in Arctic regions. Various estimates of the total amount of natural gas in place in the hydrate accumulations have been made. The most recent estimate put the range for the volume of gas trapped between 2 500 tcm and 20 000 tcm, or up to 50 times conventional reserves (see, for example, NPC, 2007). While the volume of hydrates contained in marine environments is several orders of magnitude higher than the volume in arctic permafrost regions, their concentration is higher in the latter. However, there are no commercial techniques yet available to recover them from either source. The costs could also be very high: in an offshore environment, capital and operation costs for a hydrate well could be at least twice as high as those for conventional gas (RPS, 2008). For these reasons, we do not expect commercial production of gas hydrates to begin on a significant scale before 2030. Experimental options to exploit gas hydrates include either a decrease of reservoir pressure below the level at which the gas would separate out and heating the formation to the temperature at which the same effect would occur (but the latter would require too much energy). Commercial exploitation of hydrates with minimal environmental impact requires a good understanding of the processes by which they were laid down, technologies to identify accumulations, and analysis of the interaction between their production and the rock structure.
China and India have also been conducting active hydrate research and development programmes. In 2007, a 2- to 15-metre layer of gas hydrate was discovered under the seabed of the South China Sea by the Chinese Guangzhou Marine Geological Survey. It may be possible to exploit this gas using conventional technologies, including the use of CO2 to displace methane. A major national programme in India has led to the discovery of a 130-metre thick layer of methane hydrates in the Krishna-Godavarie basin. 288
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
Arctic hydrates are mostly found beneath and within permafrost in the Canadian Arctic, the North Slope of Alaska and Siberia. Their presence offshore has been confirmed, through multi-party efforts such as the International Offshore Drilling Program, principally in the Americas, Norway, Russia and Japan (Figure 12.5). R&D on hydrates development and production has been carried out principally in Japan, Canada, Korea, the United States, China and India, and involves expenditure of some $200 million per year. In an experimental programme in the Nankai Trough, offshore Japan, measurements have been carried out to determine the properties of gas hydrates and the sediments where they are found. A joint Japanese/Canadian/US project in the MacKenzie Delta in the Canadian Arctic, launched in the late 1990s, led to the world’s first consistent and steady flow of gas following the depressurisation of gas hydrates. In March 2008, the Mallik well in northern Canada produced methane for six days, at a rate comparable to a coalbed-methane well (2 000 to 4 000 cubic metres per day).
Figure 12.5
Location of gas-hydrate resources
NW Eileen-2
Barents Sea Norwegian Sea
Mallik 2L-38 Site 889
Messoyakha Sea of Okhotsk
Blake Ridge
Hydrate Ridge
Sites 994 995 997
MITI
Site 570
The boundaries and names shown and the designations used on maps included in this publication do not imply official endorsement or acceptance by the IEA.
Gas production prospects
In the Reference Scenario, world natural gas production is projected to rise from just under 3 tcm in 2006 (and just over 3 tcm in 2007, according to provisional data) to 4.4 tcm in 2030 (Table 12.4). Production expands in all major WEO regions, with the exception of OECD Europe, where production is already in decline, and in OECD North America, where production begins to fall over the second half of the Outlook period. Output grows most strongly in both volume and percentage terms in the Middle East, tripling between 2006 and 2030 to 1 tcm. Most of this increase comes from Iran and Chapter 12 - Natural gas resources and production prospects
289
12
© OECD/IEA, 2008
Worldwide, natural gas resources are more than sufficient to meet projected demand to 2030. Nonetheless, it is uncertain whether the entire infrastructure needed to develop those resources and transport the gas from the resource-rich countries to the main centres of demand can be built in the face of economic, geopolitical and technical barriers — often referred to as “above-ground risks” (see Chapter 13). The first determinant of potential production by region is the size of the resource base and its proximity to markets, which affects the cost of supply. Gas transportation, whether by pipeline or as LNG, is very expensive when compared with oil or coal, and has become even more so in recent years with rampant cost inflation. Transport usually represents the largest share of the total cost of gas supply, especially to small consumers. Much of the world’s gas resources are located far from the main markets, so that only a small proportion can as yet be exploited profitably.
Qatar, which, alongside Russia, hold the largest reserves. Indeed, Iran sees the biggest volume increase of any country in the world (Figure 12.6). Most of the incremental output in the Middle East will be exported to North America, Europe and Asia, where indigenous output will not keep pace with demand. Production in Latin America and Africa also grows strongly. Natural gas production in the Reference Scenario (billion cubic metres) 1980
2000
2006
2015
2030
20062030*
OECD
889
1 107
1 117
1 149
1 086
-0.1%
North America
657
763
761
795
765
0.0%
78
182
188
196
164
-0.6%
554
544
524
535
515
-0.1%
219
302
305
282
217
-1.4%
Canada United States Europe Norway
26
53
89
121
127
1.5%
United Kingdom
37
115
84
44
10
-8.4%
Pacific
12
41
51
72
104
3.0%
Australia
9
33
43
64
96
3.4%
Non-OECD
634
1 425
1 842
2 363
3 348
2.5%
E. Europe/Eurasia
480
738
846
963
1 069
1.0%
Kazakhstan
n.a.
12
26
40
48
2.5%
Russia
n.a.
583
651
712
794
0.8%
Turkmenistan
n.a.
47
64
92
103
2.0%
57
247
335
449
540
2.0%
14
27
59
104
115
2.9%
Asia China India
1
25
28
41
45
2.0%
38
204
324
483
999
4.8%
Iran
4
58
104
139
313
4.7%
Qatar
3
28
49
124
169
5.3%
23
133
197
286
452
3.5%
14
86
92
106
142
1.8%
Middle East
Africa Algeria
2
13
29
64
127
6.3%
Latin America
Nigeria
36
104
139
182
287
3.1%
Argentina
10
40
45
49
60
1.1%
1
7
11
17
38
5.2%
Brazil Trinidad & Tobago Venezuela World
3
13
31
46
61
2.8%
15
28
25
33
70
4.4%
1 522
2 531
2 959
3 512
4 434
1.7%
* Average annual rate of growth.
290
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
Table 12.4
Figure 12.6
Change in natural gas production by country/region, 2007-2030
United Kingdom Canada Other OECD Europe Netherlands United States Other OECD Pacific Other E. Europe/Eurasia Uzbekistan Other Latin America India Argentina Kazakhstan Bolivia Brazil Other Africa Egypt Trinidad and Tobago Norway Mexico Turkmenistan United Arab Emirates Venezuela Libya Algeria Australia China Nigeria Qatar Saudi Arabia Other Asia Russia Other Middle East Iran –100
–50
0
50
100
150
200 250 Billion cubic metres
12
Note: Based on provisional data for 2007.
Regional trends OECD North America
7. Unless otherwise stated, all of the figures for proven reserves cited in this section are from Cedigaz (2008) and correspond to end-2007.
Chapter 12 - Natural gas resources and production prospects
291
© OECD/IEA, 2008
Canada and the United States form the largest integrated natural gas market in the world, with Canada providing about a quarter of their combined production, while there is a limited amount of trade between Mexico and the United States. Proven reserves in the entire region are currently estimated to be 8 tcm or 4.5% of the world total.7 The North American market is very mature, with the proven reserves equal to about only 10 years of production at current rates. In total, the region’s combined gas production is projected in the Reference Scenario to peak by
around the middle of the next decade and then decline gradually through to 2030. A steady rise in Mexican production is insufficient to offset declines in Canada and the United States. Canada’s production, 60% of which is exported to the United States, appears to be close to reaching a plateau. Conventional gas reserves in Western Canada represent almost a third of the remaining resource base and the National Energy Board of Canada expects this region to be the source of almost 80% of projected Canadian natural gas production in the medium term. The region also contains significant nonconventional natural gas resources, estimated at 15% of the national total, including coalbed methane, tight gas and shale gas (see section above). Further reserves in frontier areas, including the Arctic, account for more than half of Canada’s remaining ultimately recoverable gas resources. However, it is unlikely that these frontier reserves will contribute to production before 2030. Significant volumes of discovered gas are associated with offshore oil projects close to Newfoundland, but this gas is expected to be used to maintain reservoir pressure until oil production reaches an uneconomically low level.
The combination of improved technology and higher gas prices has stimulated production of deepwater and non-conventional resources, which have previously been too difficult and costly to extract. Sustained high gas prices in the last few years have resulted in higher levels of gas drilling and a trend away from drilling simple vertical wells to horizontal wells. These trends have helped to compensate for a continued decline in production from shallow-water areas in the Gulf of Mexico — output fell in 2007 for the sixth consecutive year — and falls in production in some other major producing states, including New Mexico and Louisiana. Thanks to big increases in output in Texas and Oklahoma, production in the first quarter of 2008 was 9% up on a year earlier. In the Reference Scenario, assumed high prices drive modest increases in US gas production in the medium term, but rising costs and dwindling reserves lead to a gradual decline in output over the second half of the projection period. The emergence of Alaska as a significant producing region, on the assumption that a pipeline to connect with the network in western Canada and the United States is built, is not sufficient to compensate for lower output in the lower 48 states. Nor can imports from Canada make good the remaining shortfall, so imports from outside the region (in the form of LNG) and possibly from Mexico grow through the projection period (Figure 12.7). 292
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
In the United States, proven remaining reserves of gas have increased almost every year for the last decade and, at 6 tcm, now cover almost 11 years of production at current rates. Recent drilling trends point to continued growth in reserves, with a stronger concentration on non-conventional resources, notably shales and coalbed methane. Shale formations in the lower 48 states are widely distributed, large, and contain huge resources of natural gas (as discussed in the previous section). Commercial development began only recently, yet just one field in Texas — the Barnett Shale — already contributes more than 6% of all production from the lower 48 states, a volume bigger than that from Louisiana, a traditional large producer state.
Billion cubic metres
Figure 12.7
US natural gas balance
700
Production
600
Consumption
500
Net imports
400 300 200 100 0 1990
2000
2006
2015
2030
OECD Europe European gas production peaked in 2004 at close to 330 bcm and is now in long-term decline. Continued growth in Norway has not been sufficient to offset a rapid decline in output in the United Kingdom, for many years Europe’s largest producer, despite a recent surge in drilling for gas in the North Sea. In aggregate, production in OECD Europe in the Reference Scenario is projected to decline steadily from 305 bcm in 2006 to 282 bcm in 2015 and 217 bcm in 2030 — a fall of one-third on current levels.
Large-scale extraction of oil and gas from the United Kingdom Continental Shelf began in the late 1970s. Production of both, however, is in decline and the United Kingdom became a net importer of gas on an annual basis in 2004. Output has already dropped by a third from its peak of 113 bcm in 2001. Proven reserves, at just under 650 bcm, would support production at current rates for less than 12 years. In recent years, the UK government and industry have launched a number of initiatives to help increase investment in long-term offshore projects, from improving access to pipelines and other infrastructure to ensuring activity in all areas under licence. These are proving successful. For example, the 2005 licensing round evoked more Chapter 12 - Natural gas resources and production prospects
293
12
© OECD/IEA, 2008
Norway is Europe’s leading gas producer and the sixth-largest in the world — though it holds just 3 tcm, or 1.7% of the world’s remaining proven reserves. The ultimate recoverable resource potential is thought to be large, however, as large parts of continental shelf have yet to be explored, especially the Barents Sea. The Norwegian Petroleum Directorate estimates that ultimately recoverable resources are in the range of 3.4 tcm to 6.6 tcm. However, despite their obvious potential, northern Norwegian reserves present a significant challenge, firstly due to their remoteness and the arctic conditions but also to the distance from the existing transport infrastructure and traditional markets. Production of natural gas started on the Norwegian shelf simultaneously with oil production in 1971 and exports began in 1977. Norway currently has capacity to export about 120 bcm per year, though marketed production in 2007 reached only 91 bcm, according to preliminary estimates. Output is projected to continue to grow, to over 120 bcm by 2015, and then level off at close to 130 bcm in 2030.
interest than any other round since gas production started in 1964. Monthly data shows a slowing of the rate of decline in production in early 2008. Nonetheless, increased drilling is expected only to slow, rather than halt, the long-term downward trend in production. We project output to reach 44 bcm in 2015 and then fall to only 10 bcm in 2030. The Netherlands holds the second-largest gas reserves in western Europe (1.25 tcm), the bulk of them in the Groningen gas field. Around two-thirds of its production, which amounted to 77 bcm in 2007, is exported. Proven reserves have been in gradual decline for several years: as Groningen has depleted, the portfolio of small fields has matured and discoveries have dropped off. In order to reverse this downward trend and to extend the life expectancy of the highly flexible (and valuable) Groningen field, the Dutch government has adopted several measures, including imposing a ceiling on output over discrete periods. Over the period 2006-2015, the field is allowed to produce 425 bcm. The government has also tried to stimulate exploration and production activity by improving access to information, tackling the issue of “sleeping” licences (for which no detailed development plan has been put forward) and making active approaches to exploration companies. These measures, together with higher prices have led to increased exploration, which is expected to temper the decline in output in the next few years. Nonetheless, production falls progressively more quickly through the projection period, reaching little more than 50 bcm in 2030 — a third down on 2007.
OECD Oceania Natural gas reserves in Australia and New Zealand stand at 2.5 tcm, the bulk of which are located in remote locations in offshore Western Australia. In addition to conventional natural gas, Australia has extensive deposits of coalbed methane. Australia is fast becoming an important LNG player in the Asia-Pacific region, helping to compensate for Indonesia’s declining role as an exporter of LNG. Australia’s Santos recently announced its proposed Gladstone LNG project is for a 3 to 4 million tonnes per year LNG processing train and associated infrastructure, based on coalbed methane from Queensland’s Bowen and Surat Basins, the first of its kind. The project aims to export its first shipment by 2014. Total Australian production is projected to double by 2030, to close to 100 bcm from less than 50 bcm in 2007. New Zealand production is expected to remain modest.
Russia has 48 tcm of proven gas reserves, one-quarter of the world total and the largest share of any country. The USGS estimates undiscovered recoverable conventional resources, not including reserves growth, at 80 tcm, including an estimated 37 tcm located north of the Arctic Circle (USGS, 2000 and 2008). Russia is the world’s largest natural gas producer and exporter, and is the second-largest natural gas consumer after the United States. Russian exports meet almost a quarter of OECD Europe’s gas needs, directly to Finland and Turkey, and via transit pipelines through Ukraine or Belarus. 294
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
Eastern Europe/Eurasia
Russia’s resources are unquestionably large enough to support continuing long-term growth in production. But there are doubts about the cost of developing those resources and the speed with which Gazprom, the dominant gas producer, and independent companies can proceed. Gazprom reported reserves replacements in 2007 of 585 bcm — 6% more than it produced and the third year in a row in which it was able to replace all its production. But during the previous ten years, the reserves replacement ratio was only around 50% to 70%. The Russian gas industry is at a cross-roads as production shifts from the super-giant fields that have accounted for most of the country’s output for decades — notably Urengoye, Yamburg, Medvezhye and more recently Zapolyarnoye — to fields located in new, more difficult-to-develop regions, needing new transportation infrastructure. Gazprom has begun development work on the 4-tcm Bovanenkovskoye field, the first to be brought into production on the Yamal peninsula in northern Siberia. Production is due to begin around 2012, with output expected to reach 115 bcm per year once a new pipeline is built to link the field to the national transmission system. Other Yamal fields are lined up for development. Yamal presents major technical challenges because of its extreme climate and permafrost, making it difficult to stabilise infrastructure. The annual operational window for getting equipment to the construction site is also very tight. Furthermore, new warmer climate trends are affecting construction plans, often requiring new project design, which needs to be tested before being put in place, causing delays to project timelines. Given these difficulties and the scale of the project, many industry observers believe that the project may not be commissioned on time. Gazprom, in partnership with Total and StatoilHydro, also plans to bring the 3.8-tcm Shtokman field, which is located offshore in the Barents Sea, into production by 2012. Production is planned to reach 40 bcm per year at plateau in the first two phases of the field’s development and 70 bcm after one or two more phases. Development of this field also faces major technical hurdles, as the field is located 550 kilometres offshore in water prone to ice-floes. A long-distance pipeline to European markets will also need to be built. Total development costs have ballooned to more than $40 billion for the first two phases. We assume that first gas comes after 2015.
12
Caspian region producers are expected to make a major contribution to the growth in gas production in the Eastern Europe/Eurasia region (Figure 12.8). The contribution of the Caspian region to global gas supply will depend on the level of investment in Chapter 12 - Natural gas resources and production prospects
295
© OECD/IEA, 2008
In the Reference Scenario, Russian gas production is projected to rise from 680 bcm in 2007 to about 710 bcm in 2015 and just under 800 bcm in 2030 — an overall increase of just under one-fifth. There are enormous uncertainties about the mobilisation of the massive investments needed to develop these new fields and build pipelines to connect them to the domestic network and export systems. Gazprom has increased its capital spending programme for 2008 to around $35 billion, of which about $22 billion is earmarked for the upstream sector and pipelines. It plans to invest a total of $140 billion over the five years to 2012. A growing share of gas investment in Russia is expected to come from independent gas producers, on the assumption that their access to Gazprom’s transmission system is made easier. Gazprom will nonetheless remain the dominant producer for the foreseeable future.
Billion cubic metres
Figure 12.8
Natural gas production in Eastern Europe/Eurasia
1 200
Other
1 000
Kazakhstan Turkmenistan
800
Russia
600 400 200 0 1990
2000
2006
2015
2030
exploration and production, and on the availability on commercial terms of reliable routes to international markets. Volumes available for export will also depend on the region’s own demand for gas, which has been growing strongly on the back of heavy subsidies; gas is used very inefficiently across the region.
Uzbekistan is a major gas producer, with production of 65 bcm in 2007, of which 11 bcm is exported, mainly to Russia. Following signs of a slowdown in oil and gas output in 2005, Uzbekistan has sought to increase investment in exploration and production to stem the decline in output at its mature fields and has concluded a number of new production-sharing agreements, predominantly with Russian and Asian companies. Output is projected to remain broadly flat through to 2030. Russia is likely to remain the most significant export market, but the construction of a gas pipeline from Turkmenistan through Uzbekistan (and Kazakhstan) to China will open up the possibility of gas trade with China when it is completed in 2011 (though most of the gas that will flow through the line is likely to come from Turkmenistan). 296
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
Turkmenistan has ambitious plans to increase production and exports. Turkmenistan is the largest producer in Central Asia, with output of 72 bcm in 2007; its relatively small population (4.9 million) means that large volumes are available for export. Russia is currently the main importer of Turkmen gas, with smaller volumes going to Iran. Estimates of Turkmenistan’s natural gas reserves vary considerably, introducing a large element of uncertainty into projections of its future contribution to global gas supply. Official sources have put gas reserves in the country at more than 20 tcm, an amount approaching the range of proven reserves in Iran or Qatar — and far more than the 3 tcm of proven reserves reported elsewhere (for example, BP, 2008 and O&GJ, 2007). The government announced in April 2008 an international audit of Turkmenistan’s gas reserves. The Turkmen government aims to raise production to 250 bcm per year by 2030, but there is considerable uncertainty that the known resource base would support such an increase or that sufficient investment will be forthcoming. Major investment is required to develop new reserves to compensate for declining output from the mature fields that provide most of the country’s output now. We project annual production to rise to over 100 bcm by 2030.
Production is projected to increase moderately in Kazakhstan. Almost all of the gas produced there is associated with oil production, notably from the Tengiz and Karachaganak projects in the west of the country. Much of this gas is re-injected, in order to maintain reservoir pressure and enhance oil output, or flared for lack of infrastructure to transport it to market. Most of the gas produced in Kazakhstan is consumed domestically. Annual gas production is projected to increase from 26 bcm in 2006 to around 40 bcm by 2015 and close to 50 bcm by 2030. Gas production in Azerbaijan is also set to grow, with the offshore Shah Deniz field making a sizable contribution, providing up to 20 bcm per year for export to Turkey and other European countries from the middle of the next decade.
Non-OECD Asia Asia is a small producer and consumer of gas relative to the size of its population and its overall energy use. The region as a whole is a net exporter, though some countries — including China and India — need to import some or all of the gas they consume. Asian gas production — most of which is consumed locally — amounted to 341 bcm in 2007 (little more than half that of Russia), having grown by 95 bcm, or 39%, since 2000. Annual output rates are projected to expand further, to around 450 bcm in 2015 and 530 bcm in 2030. Indonesia has the largest proven gas reserves in the region, estimated at 3 tcm, and is the largest producer. Reserves grew sharply in 2006 and 2007, with new discoveries (resulting from a spurt in the issue of new exploration licences) and increased appraisal drilling. Production has nonetheless continued to drift lower as output from the main fields, which have been producing for many years, declines. Total output stood at 70 bcm in 2007. Indonesia’s difficulties in meeting its contractual commitments and extending existing contracts as they expire have been an important factor in the run-up in LNG prices in the Pacific region in recent years. There is a major geographical mismatch between the main demand centres in Indonesia, in Java and Bali, and the main reserves. Pipelines connect these demand centres to procuring areas on Natuna Island and South Sumatra, but other resource-rich areas, such as Kalimantan and Papua, are not connected because of their remoteness. Due to this mismatch, LNG import terminals are being considered in East Java, West Java and North Sumatra.
12
China’s proven reserves of natural gas amount to 2.7 tcm, about a fifth of which are associated with oil and 90% located onshore. Most production to date has come from onshore associated reserves; onshore non-associated gas reserves are largely untapped. Chapter 12 - Natural gas resources and production prospects
297
© OECD/IEA, 2008
Exports to Southeast Asia and further afield, together with rising domestic demand, are expected to stimulate a modest rebound in Indonesian production through to 2030. There are already pipelines to its neighbours, including from West Natuna to Singapore and Malaysia, and from South Sumatra to Singapore. The government plans to make more gas available to the domestic market to replace oil, which will drive up demand (especially in power generation). It hopes that moves to speed up the approval process for new upstream projects and more transparent tenders will encourage greater participation by international and domestic players in exploration and appraisal activities. We project Indonesia’s production to rise to 90 bcm in 2030.
Production has been rising swiftly in recent years, reaching an estimated 68 bcm in 2007 — more than 60% higher than in 2004. Only a small number of gas fields have reached their production peak and several fields with large reserves have yet to be developed, so there is significant potential to boost output in the short to medium term. In the Reference Scenario, China’s annual production is projected to reach more than 100 bcm by 2015 and 115 bcm by 2030, by which time output is expected to be in decline.
Middle East The Middle East holds around 40% of world gas reserves with Iran, Qatar, Saudi Arabia and United Arab Emirates holding the largest volumes. Gas reserves in Iran and Qatar alone account for nearly 30% of the world total. Yet, despite a rising trend in production in recent years, these two countries together account for only 5% of global production and the region as a whole for less than 11%, suggesting strong potential to boost supply in the long term, particularly in Iran. Production in the Middle East is expected to grow more than in any other region. In the Reference Scenario, we project Middle East output will jump from 318 bcm in 2007 to 480 bcm in 2015 and to 1 tcm in 2030. Iran remains the region’s biggest gas-producing country in the Reference Scenario, its production rising from an estimated 107 bcm in 2007 to over 350 bcm in 2030 — on the assumption that the required investment can be mobilised. The lion’s share of the country’s reserves, and the expected source of much of the production growth in the medium term, is in the Iranian part (South Pars) of the South Pars/North Dome gas/ condensate field. The combined structure is the world’s largest gasfield, with 40 to 50 tcm of reserves. More than one-third of these reserves are in the Iranian sector. South Pars currently accounts for about one-quarter of Iran’s total gas production and just under half of the country’s 28 tcm of proven reserves.
Increasing domestic demand for re-injection, power generation and end uses, however, is likely to mean further delays in planned gas-export initiatives. In mid-2008, Iran officially postponed two of the three LNG developments at South Pars planned for the 2010-2013 period, leaving only the domestic LNG development in place. Shell and Total are now discussing the prospect of LNG development linked to later phases of the South Pars field. China’s CNOOC and Malaysia’s SKS have held intermittent talks on LNG development of other fields, but no firm investment commitments have been made as 298
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
The government has, for many years, had ambitious plans to boost production from South Pars and other fields to supply the rapidly growing domestic market and to export (by pipeline and as LNG), but developments have been delayed for technical and political reasons. New upstream awards have slowed markedly since 2004, due to changes in political priorities and the international isolation resulting from trade embargoes imposed in response to the country’s nuclear programme. A growing proportion of wellhead gas supplies has been re-injected to maintain oil production in some old fields which are faced with high natural decline rates (see Chapter 10). As a result, production growth is likely to slow significantly through to the middle of the next decade. Boosting capacity thereafter will depend on a resurgence in awards and the clearance of other obstacles in the next few years. A large-scale internal pipeline construction programme is underway to accommodate higher production to meet domestic demand.
yet. In addition, prolonged negotiations about gas deliveries to the Iran-Pakistan-India (IPI) pipeline mean that the start date for this project is now 2013 or beyond, rather than the original date of 2011. Qatar, with proven reserves of 26 tcm, is far and away the biggest gas exporter in the Middle East and the region’s fastest-growing producer. We project its production to rise from 54 bcm in 2007 to almost 170 bcm in 2030, driven mainly by LNG exports. Projects currently being developed by Qatar Petroleum with foreign partners will boost the country’s LNG export capacity to 105 bcm per year by 2011, up from 40 bcm at end2007 (Table 12.5). Qatar became the world’s biggest LNG exporter in 2006 and looks set to retain that title well into the future. The world’s largest trains, of 10.6 bcm per year each, are under construction in the Qatargas and Rasgas ventures, as well as the world’s largest gas-to-liquids plant at the Shell/Qatar Petroleum Pearl project (70 kb/d). The world’s biggest GTL plant, Oryx, with a capacity of 34 kb/d, was commissioned in Qatar in 2006. The current raft of projects, however, has been hit by technical problems and cost escalation, resulting from difficulties in securing materials and services. This has contributed to delays of more than six months in the commissioning of the first train at Qatargas 2 and the postponement of Rasgas 3 to early 2009 (delays which are by no means unique to Qatar). Project participants believe that this will have a knock-on effect through the project chain. Table 12.5
Qatari LNG projects
Trains
Project
Start date (previous target)
bcm per year
Qatargas
Partners
55.4
1-3
1997-1998
12.9
QP, ExxonMobil, Total, Marubeni, Mitusi
Qatargas 2
Q2 2008 (Q1)
10.6
QP, ExxonMobil
2009 (2008)
10.6
QP, ExxonMobil, Total
6
Qatargas 3
2010 (2009)
10.6
QP, ConocoPhillips, Mitsui
7
Qatargas 4
2011 (2010)
10.6
QP, Shell
4 5
RasGas
49.4
1-2 3
RasGas 2
1999
9.0
QP, ExxonMobil, Kogas, Itochu, LNGJapan
2004
6.4
QP, ExxonMobil
4
2005
6.4
QP, ExxonMobil
5
2007
6.4
QP, ExxonMobil
2009 (2008)
10.6
QP, ExxonMobil
10.6
QP, ExxonMobil
6
12
RasGas 3
7
2010 (2009)
Total capacity in 2011
104.8
The prospects for further growth in Qatari gas production beyond 2012 are clouded by the uncertainty created by a moratorium on new export projects, which was imposed in 2005 while the effect of existing projects on North Field reservoirs was studied. Chapter 12 - Natural gas resources and production prospects
299
© OECD/IEA, 2008
Source: IEA (2008).
Rather than make early commitments further to boost exports, in the form of LNG, GTL or pipeline gas (gas is already exported to the United Arab Emirates through the Dolphin pipeline), Qatar may seek first to prove up more reserves through additional exploration and appraisal, perhaps first at greater reservoir depths in the North Field. The government is also concerned about ensuring adequate supplies to domestic users, including a burgeoning petrochemical sector. In 2007, gas blocks at the North Field assigned to a cancelled GTL project (a joint venture between Qatar Petroleum and ExxonMobil) were reassigned to the domestic Barzan project, which will supply the domestic industrial sector. Gas production in the United Arab Emirates is projected to grow less rapidly than in neighbouring states, from 46 bcm in 2007 to 72 bcm in 2030. Surging energy demand from industrial projects, power generation and desalination plants has led to shortages in the Emirates of Abu Dhabi and Dubai. Although reserves are large, totalling over 6 tcm, most existing production is earmarked for re-injection at oilfields and for LNG exports from the country’s only plant, at Das Island in Abu Dhabi. Rising international gas prices have made it more attractive to develop the country’s reserves, but gasquality problems are holding back investment. For example, although two sour gas fields at Shah and Bab have been put out to tender for development with foreign contractors, negotiations on the Shah field have been hampered by high costs — estimated at between $4 and $5 per MBtu — because of the difficulties in accessing the reserves and the high sulphur content. Bab, together with an integrated gas project and the sour gas development at the Hail field, are expected to provide additional volumes, but much of these will be required for re-injection to boost production at the country’s ageing oilfields. In Saudi Arabia, Saudi Aramco’s upstream exploration programme has continued to prioritise exploration for non-associated gas. Most of the 67 bcm of gas it produced in 2007 was associated with oil. It has laid out plans to increase reserves, which now stand at 6 tcm, by an average 142 bcm per year over the next ten years. Some successes have been reported, although efforts with foreign partners in the Empty Quarter have not yet yielded any new reserves. Development plans are currently focused on the offshore Karan field, which is now due to supply some 16 bcm per year into the national gasgathering system from the end of 2011 — 5 bcm per year more than originally planned. The switch to fuel oil and crude oil in some power stations will make more gas available for domestic uses in the medium term. However, Saudi Arabia remains unlikely to consider any export of natural gas until it is certain that its own increasing needs are assured. We project Saudi gas production to reach almost 190 bcm in 2030.
Three countries — Algeria, Egypt and Nigeria — account for over 90% of all the gas produced in Africa. In aggregate, production is projected to more than double from 206 bcm in 2007 to over 450 bcm in 2030 in the Reference Scenario — the largest increase in volume terms of any region bar the Middle East. The three main producers account for the bulk of this increase in volumetric terms, but the share of other countries, notably Libya and Equatorial Guinea, grows over the projection period. 300
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
Africa
Nigeria is expected to overtake Algeria to become the region’s biggest producer by 2030. With an estimated 5.2 tcm of proven natural gas reserves — the fifth largest in the world and the largest in Africa — output is projected to reach close to 130 bcm in 2030, up from just over 35 bcm in 2007. More than half of the country’s marketed output is exported as LNG, with the remainder flared (see Chapter 15) or used for power generation and in industry (there is no public distribution network for the residential sector). A large share of total output is used for oil extraction. Future production growth will be driven by both export projects and rising domestic demand, particularly for power generation. With the launch of Train 6 in December 2007, the Nigeria LNG (NLNG) project, located on Bonny Island in the Niger Delta, reached a total LNG production capacity of 31 bcm per year, overtaking Algeria and becoming the largest-capacity LNG facility in the Atlantic Basin. NLNG is investigating the possibility of adding huge seventh and eighth trains, of 10.9 bcm per year each, though a final investment decision has yet to be made. Other projects are planned. In total, capacity could reach over 60 bcm per year by 2015. Nigeria has also developed projects to export natural gas via pipelines. The West Africa Gas Pipeline project will carry natural gas from Nigeria to Benin, Togo and Ghana. The 678-km line was completed in December 2007 but commercial operation is not expected to begin until 2008, a delay of more than one year from the original target date at the time of the construction decision.
Egypt, with 2.1 tcm of proven gas reserves, raised production from 22 bcm in 2000 to 52 bcm in 2007. Most of the increase has gone to the domestic market, but exports have grown too, quadrupling between 2004 and 2007, following the opening of the Arab gas pipeline to Jordan and the commissioning of three LNG trains. Egypt does not permit export commitments to exceed one-third of the reserve base, in order to ensure adequate domestic supplies for future generations. Two new export outlets were opened up in early 2008, in the form of a pipeline to Israel and the extension of the Arab gas pipeline from Egypt through Jordan into Syria. That pipeline will be linked Chapter 12 - Natural gas resources and production prospects
301
12
© OECD/IEA, 2008
Algeria, with 4.5 tcm of proven gas reserves, has the potential to continue to expand its production to meet growing domestic demand, particularly for power generation and water desalination, and for export. Output is projected to rise from almost 90 bcm in 2007 to over 140 bcm per year by 2030. The state-owned company, Sonatrach, which dominates the industry, aims to achieve gas exports (by pipeline to Europe and as LNG) of 85 bcm in 2012 and 100 bcm in 2015, up from 62 bcm in 2007. Gas reserves in the Berkine Basin, as well as more difficult reserves from the southwest of the country, are expected to feed these projects. They will require large infrastructure investments around Hassi R’Mel, the country’s largest single gasfield. Further supplies are envisaged from recent exploration successes, which included 20 oil and gas discoveries in 2007. Algeria resumed licensing rounds for upstream acreage in 2008, after a hiatus caused by revisions to the hydrocarbon law in 2005 and 2006. Sonatrach is negotiating partnerships in upstream projects with companies in importing countries in Europe, in return for direct access to those companies’ domestic markets. Alliances have also been pursued with other producers, including Norway’s StatoilHydro, with an eye to joint developments, cargo swaps and market access (on this occasion) to StatoilHydro’s capacity at the US Cove Point terminal.
into the Turkish grid in the next phase of development, although there will be limited volumes of gas available for onward delivery in the near term. A final investment decision is still pending on a second train at the Damietta LNG plant. Other countries are set to emerge as major producers and exporters. In Libya, the emphasis is on export markets: domestic consumption is expected to increase rapidly, but from a small base, due to the limited population and small industrial base. A 2004 deal with Shell could lead to a new LNG project, if sufficient reserves were to be found. BP has also undertaken exploration commitments, with a view to locating feedgas for an LNG facility. ExxonMobil is actively searching for potentially exportable gas offshore. At the end of 2007, the only current gas exporter, Eni, committed to a ten-year deal which includes the further development of its Libyan fields to supply an additional 3 bcm of gas through the 8 bcm Greenstream pipeline to Italy and construction of a new 5 bcm per year LNG export plant at Mellitah. The LNG deals are unlikely to yield significant gas exports before 2013, but do make Libya an expected player in regional export growth in the latter half of the next decade. The country held its first upstream licensing round focused specifically on gas in 2007, with a view to more than doubling its existing proven reserve base of 1.3 tcm. Some other African countries are also aiming to develop LNG projects. Angola, with large associated gas reserves, is building its first LNG liquefaction plant in the Zaire province with capacity of 7 bcm per year. It is expected to be completed by early 2012. Equatorial Guinea began exporting LNG in 2007 from a 4 bcm per year plant.
Latin America
All the main gas-resource holders are expected to see rapid growth in production. Venezuela, which has enough gas resources to overtake Argentina as the leading producer in Latin America, has proven reserves of 5.1 tcm accounting for two-thirds of the total for the whole continent. We project its output to jump from only 25 bcm in 2007 to 70 bcm in 2030, though the above-ground risks to this projection are high. Exports of LNG are likely to drive much of the growth in the latter part of the Outlook period. Brazil’s output, all of which will be needed to meet domestic demand, is projected to more than triple, to 38 bcm by 2030, with a sharp increase in output expected to come from recent offshore pre-salt finds (see Chapter 9). 302
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
Gas production and consumption in Latin America remain highly concentrated. Argentina, Bolivia, Brazil, Colombia, Trinidad and Tobago, and Venezuela account for almost all the region’s production, most of which (with the exception of that from Trinidad and Tobago, which exports LNG) is consumed within the domestic market. Although the region has plentiful gas reserves, estimated at 7.7 tcm, the recent surge in producing countries of resource nationalism and government policies that discourage upstream investment has led some importing countries to turn to imported LNG from outside the region, rather than rely on overland pipeline imports from neighbouring countries. Today, several LNG receiving terminals are either being built or are at the planning stage — three in Brazil, two in Chile, one in Argentina and one in Uruguay. If all are built, Latin America could be importing as much as 14 bcm per year of LNG by the end of the decade, reaching 27 bcm per year by the end 2012.
CHAPTER 13
UPSTREAM INVESTMENT PROSPECTS Are we spending enough? H
I
G
H
L
I
G
H
T
S
Worldwide investment in the upstream oil and gas industry has surged in nominal terms in recent years, partly due to soaring costs and higher oil prices, which have strengthened incentives to explore for and develop reserves in some regions. Total upstream investment more than tripled between 2000 and 2007, from an estimated $120 billion to $390 billion. Most of this increase was due to higher unit costs: in cost-inflation adjusted terms, investment in 2007 was about 70% higher than in 2000. Based on the plans of the companies surveyed for this Outlook, global upstream oil and gas investment is expected to rise to just over $600 billion in nominal terms by 2012 — an increase of more than half over 2007. If costs level off, as assumed, real spending in the five years to 2012 would grow by 9% per year — about the same rate as in the previous seven years. Spending on exploration will remain small compared with that on field development. The bulk of upstream investment will continue to be made by the international companies. Over the period 2000-2007, the five super-majors alone accounted for 29% of all the upstream investment made by the companies surveyed for this Outlook and other international and private upstream companies for a further 31%. These shares will not change much over the next five years or so. Much of the increase in this upstream spending will go to meeting rising unit costs of developing dwindling resources in non-OPEC regions, where the bulk of their investment is directed.
Current projects are expected to add more than enough capacity to meet demand growth and counter declines at existing fields through to 2010. But many more projects will need to be sanctioned (mostly within the next two years) to bridge the gap between the total gross capacity addition of 30 mb/d needed by 2015 and the actual projected addition of 23 mb/d from current projects. If actual capacity additions fall short, spare production capacity would diminish and oil prices would undoubtedly rise.
Chapter 13 - Upstream investment prospects
303
© OECD/IEA, 2008
Worldwide, upstream unit costs rose on average by an estimated 90% between 2000 and 2007 and by a further 5% in the first half of 2008, based on the IEA Index of Upstream Capital Costs. Most of the increase occurred in 2004-2007. Exploration and development costs have risen at roughly the same rates. Taking account of these cost increases, overall physical upstream activity over the seven years to 2007 grew by about 70%, or 8% per year.
The Reference Scenario projections imply a need for cumulative investment in the upstream oil and gas sector of around $8.4 trillion (in year-2007 dollars) over 2007-2030, or $350 billion per year. That is significantly less than is currently being spent. But there is a major shift in where that investment is needed. Much more capital needs to go to the resource-rich regions, notably the Middle East, where unit costs are lowest. It is far from certain that these countries will be willing to make this investment themselves or accept sufficient foreign investment to make it up.
Recent investment trends and near-term outlook Worldwide investment in upstream oil and gas has surged in nominal terms in recent years and, on current plans, is set to rise further in the coming years. This reflects a combination of higher oil prices, which have strengthened incentives for oil and gas companies to explore for and develop reserves in some regions, soaring cash-flow and rising costs. The prices of cement, steel and other materials used in building production facilities, the cost of fuel used in construction, the cost of hiring skilled personnel and drilling rigs, and the prices of oilfield equipment and services have soared, especially since 2003, though there are signs that costs may now be levelling off. Between 2000 and 2007, capital spending grew at an average rate of 18% per year. In 2007, total upstream investment reached an estimated $390 billion, up from $120 billion in the year 2000 — a more than three-fold nominal increase. The increase has been particularly strong since 2004 (Figure 13.1). This trend is broadly in line with data reported by other organisations. Most of the increase in spending since 2000 was needed to meet higher unit costs: in cost-inflation adjusted terms, investment in 2007 was 70% higher than that in 2000. Box 13.1 describes the methodology used to analyse investment trends.
Billion dollars
Figure 13.1
World investment in oil and gas exploration and production
700
IEA survey
600
IEA (estimated world) Lehman Brothers
500 400 300 200
0 2000
2002
2004
2006
2008
2010
2012
Sources: IEA databases and analysis; Lehman Brothers (2007).
304
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
100
Based on the announced investment plans or forecasts by the 50 companies surveyed for this Outlook, total upstream oil and gas investment is expected to rise to just over $600 billion in nominal terms by 2012 — an increase of more than half over 2007. If this expectation is realised, annual investment in nominal terms would be on average two-and-a-half times higher in 2008-2012 than in 2000-2007. Upstream investment as a share of total oil and gas industry investment (including refining, oil pipelines, liquefied natural gas chains and gas-to-liquids) is expected to rise slightly from 68% in 2000-2007 to 70% in 2008-2012. In cost-inflation adjusted terms, upstream spending is set to grow by around 50% between 2007 and 2012, assuming unit costs level off in 2008. This equates to 9% per year. There is little scope for the industry to increase real spending faster than this, even if major new investment opportunities arise, because of a lack of skilled labour and equipment. These investment trends should be treated as indicative only, as they are based on current plans, which could change were oil prices and costs to differ markedly from our assumptions. Box 13.1
Approach to assessing the outlook for upstream investment
Our outlook for upstream investment is assessed both bottom-up and top-down. The former involves a detailed analysis of the plans and prospects for oil and gas industry investment over the period 2000 to 2015, with the aim of determining whether the industry is planning to invest more in response to higher prices and to a growing need for new capacity and of assessing the resulting additions to production capacity. This analysis is based on two main elements: A survey of the capital-spending programmes of 50 of the largest upstream oil and gas companies (national and international companies and pure exploration and production companies), covering actual capital spending from 2000 to 2007 and their plans or forecasts of spending through to 2012 (Table 13.1). Companies were selected on the basis of their size as measured by their production and reserves, though geographical spread and data availability also played a role. The surveyed companies account for close to 80% of world oil production and 79% of oil reserves, 66% of gas production and 71% of gas reserves. Total industry investment was calculated by adjusting upwards the spending of the 50 companies, according to their share of world oil and gas production for each year.
13
Data was obtained from companies’ annual and financial reports, corporate presentations, press reports, trade publications and direct contacts in the industry.
Chapter 13 - Upstream investment prospects
305
© OECD/IEA, 2008
A review of all major upstream projects worldwide — sanctioned (approved by the company board) and planned — which are due to be brought on stream before 2015. This review covers over 570 projects, each involving a capacity addition of more than 5 thousand barrels of oil per day (kb/d). Projects include conventional oil and gas production, non-conventional oil sands and coal- and gas-to-liquids.
The top-down determination of overall upstream investment needs to 2030 was derived from the WEO oil- and gas-supply models. These models determine the rate of development of new projects, based on assumptions — derived from our review of upstream projects — about unit capital costs which take into account the impact of technological change and country-specific factors. The results are presented in year-2007 dollars. With this approach, capital spending is attributed to the year in which the plant in question becomes operational, because of the difficulties in estimating lead-times and how they might evolve in the future. For both the near-term and longer-term assessments, investment is defined as capital expenditure only. It does not include spending that is usually classified as operation and maintenance.
The bulk of upstream investment will continue to be made by the international companies. Over the period 2000-2007, the five super-majors1 alone accounted for 29% of all the upstream investment of the companies surveyed for this Outlook and other international and private upstream companies for a further 31% (Figure 13.2). Their combined share is projected to drop by just one percentage point to 59% over 20082012. The share of national companies is set to rise from 40% to 41%. More than half of total upstream investment over the next five years will, as before, continue to go to maintaining output at oil and gas fields already in production. Spending on exploration is expected to remain small compared with that on the development of existing and new fields (Figure 13.3). Figure 13.2
Upstream investment of surveyed companies by type of company
2000-2007
Super-majors
2008-2012
14%
IOCs
14%
NOCs 40%
29%
27% 41%
17%
E&P
18%
1. ExxonMobil, Shell, BP, Total and Chevron.
306
World Energy Outlook 2008 - OIL AND GAS PRODUCTION PROSPECTS
© OECD/IEA, 2008
Note: For 50 surveyed companies. IOC = international oil company; NOC = national oil company. E&P = exploration and production; breakdown by type of company. Source: IEA database and analysis.
Companies in investment survey, 2007
Company
Type
Oil production (mb/d)
Saudi Aramco
NOC
10.9
Gas production (bcm) 68.4
NIOC
NOC
4.4
106.7
Pemex
NOC
3.5
56.0
Adnoc
NOC
2.7
22.5
ExxonMobil
Super-major
2.6
97.0
KPC
NOC
2.2
12.0
BP
Super-major
2.4
84.2
PDVSA
NOC
2.6
25.0
CNPC
NOC
2.8
57.8
Rosneft
NOC
2.0
15.7
INOC
NOC
2.3
4.3
Lukoil
IOC
2.0
14.0
Shell
Super-major
1.9
84.9
Petrobras
NOC
1.9
23.7
LNOC
NOC
1.9
10.9
Chevron
Super-major
1.8
51.9
Sonatrach
NOC
1.6
90.0
Total
Super-major
1.5
50.0
TNK-BP
IOC
1.4
8.5
NNPC
NOC
1.4
18.3
StatoilHydro
NOC
1.1
34.7
ConocoPhillips
IOC
1.1
52.6
Eni
IOC
1.0
42.3
Qatar Petroleum
NOC
1.0
42.9
Gazprom
NOC
0.9
548.5
Petronas
NOC
0.8
57.3
Sinopec
NOC
0.8
8.0
Sonangol
NOC
0.7
0.8
ONGC
NOC
0.5
22.3
Repsol YPF
IOC
0.5
32.3
Occidental
E&P
0.4
7.3
CNOOC
NOC
0.4
5.8
Anadarko
E&P
0.3
19.7
Apache
E&P
0.3
18.6
CNR
IOC
0.3
14.0
Hess
E&P
0.3
6.9
PetroCanada
IOC
0.3
6.1
BG
IOC
0.2
26.6
Devon
E&P
0.2
24.5
Talisman
E&P
0.2
10.5
Marathon
IOC
0.2
9.6
Chapter 13 - Upstream investment prospects
13
307
© OECD/IEA, 2008
Table 13.1
Table 13.1
Companies in investment survey, 2007 (continued)
Company
Type
Nexen
E&P
Oil production (mb/d) 0.2
2.0
EnCana
E&P
0.1
37.7
XTO
E&P
0.1
15.0
Pertamina
NOC
0.1
10.6
PTT
NOC
0.1
9.6
Woodside
E&P
0.1
5.9
Plains
E&P
0.1
0.8
Pioneer
E&P